Поиск патентов

Настройки

Глубина выборки

Укажите год

Небесная энциклопедия

Космические корабли и станции, автоматические КА и методы их проектирования, бортовые комплексы управления, системы и средства жизнеобеспечения, особенности технологии производства ракетно-космических систем

Подробнее

Мониторинг СМИ

Мониторинг СМИ и социальных сетей. Сканирование интернета, новостных сайтов, специализированных контентных площадок на базе мессенджеров. Гибкие настройки фильтров и первоначальных источников.

Подробнее

Форма поиска

Ключевые слова. Может быть несколько по одной на строку

Поддерживает ввод нескольких поисковых фраз (по одной на строку). При поиске обеспечивает поддержку морфологии русского и английского языка

Автор

Ведите корректный номера.

Владелец

Ведите корректный номера.

Классы IPC

Ведите корректный номера.

Классы CPC

Ведите корректный номера.

Начиная с года

Укажите год

Заканчивая годом

Укажите год

Применить Всего найдено 3005. Отображено 100.

31-05-2012 дата публикации

System and method for detecting global harmful video

Номер: US20120134578A1

Автор: Byeong Cheol Choi, Byung Ho Chung, Hyun Sook Cho, Jae Deok Lim, Seung Wan Han

Принадлежит: Electronics and Telecommunications Research Institute ETRI

A system for detecting a global harmful video includes: a video determination policy generation unit for determining harmfulness of learning video segments from video learning information to analyze occurrence information of harmful learning video segments, and generating a global harmfulness determination policy based on the occurrence information; and a video determination policy execution unit for determining harmfulness of input video segments from information of an input video to analyze occurrence information of harmful input video segments, and determining whether the input video is harmful or not based on the occurrence information of the harmful input video segments and the global harmfulness determination policy.

Подробнее

Номер записи: 1

13-09-2012 дата публикации

Computer-readable storage medium storing image processing program, image processing method, and image processing device

Номер: US20120230589A1

Автор: Daiki Masumoto, Masahiko Sugimura, Masaki Ishihara, Shigemi Nagata, Shugo Nakamura, Susumu Endo, Takayuki Baba, Yusuke Uehara

Принадлежит: Fujitsu Ltd

A computer-readable storage medium storing an image processing program that causes a computer to execute a process includes acquiring the same object identification information included in a plurality of image data items by referring to a storage unit that stores each of the image data items, object identification information that identifies an object included in the image data item, and a location information item that identifies a location of the object in the image data item in association with one another; acquiring the location of the object identified by the acquired object identification information in each of the image data items by referring to the storage unit; computing a difference between the acquired location information items; comparing the difference between the location information items with a predetermined location threshold value; and determining whether the image data items are to be in the same group.

Подробнее

Номер записи: 2

06-12-2012 дата публикации

Online environment mapping

Номер: US20120306847A1

Автор: Jan-Michael Frahm, Jongwoo Lim, Marc Pollefeys

Принадлежит: Honda Motor Co Ltd

A system and method are disclosed for online mapping of large-scale environments using a hybrid representation of a metric Euclidean environment map and a topological map. The system includes a scene module, a location recognition module, a local adjustment module and a global adjustment module. The scene flow module is for detecting and tracking video features of the frames of an input video sequence. The scene flow module is also configured to identify multiple keyframes of the input video sequence and add the identified keyframes into an initial environment map of the input video sequence. The location recognition module is for detecting loop closures in the environment map. The local adjustment module enforces local metric properties of the keyframes in the environment map, and the global adjustment module is for optimizing the entire environment map subject to global metric properties of the keyframes in the keyframe pose graph.

Подробнее

Номер записи: 3

03-01-2013 дата публикации

Video dna (vdna) method and system for multi-dimensional content matching

Номер: US20130006951A1

Автор: Lei Yu, Yangbin Wang, Yichao Zhang

Принадлежит: Individual

A method and system of identifying and matching content characteristics comprises the steps of ingesting VDNA (Video DNA) fingerprints from input media contents, quick hash-based query across the VDNA registered indexer servers, and performing multi-dimensional content identification in query engines to obtain best matched results of the input media content.

Подробнее

Номер записи: 4

18-07-2013 дата публикации

Feeling-expressing-word processing device, feeling-expressing-word processing method, and feeling-expressing-word processing program

Номер: US20130182907A1

Автор: Kyota Higa, Masumi Ishikawa, Toshiyuki Nomura, Yuzo Senda

Принадлежит: NEC Corp

The present approach enables an impression of the atmosphere of a scene or an object present in the scene at the time of photography to be pictured in a person's mind as though the person were actually at the photographed scene. A feeling-expressing-word processing device has: a feeling information calculating unit 11 for analyzing a photographed image, and calculating feeling information which indicates a temporal change in a scene shown in the photographed image or a movement of an object present in the scene; and a feeling-expressing-word extracting unit 12 for extracting, from among feeling-expressing words which express feelings and are stored in a feeling-expressing-word database 21 in association with the feeling information, a feeling-expressing word which corresponds to the feeling information calculated by the feeling information calculating unit 11.

Подробнее

Номер записи: 5

06-01-2022 дата публикации

METADATA GENERATION FOR VIDEO INDEXING

Номер: US20220004574A1

Автор: Figov Zvi, Ofer Irit

Принадлежит:

A video indexing system identifies groups of frames within a video frame sequence captured by a static camera during a same scene. Context metadata is generated for each frame in each group based on an analysis of fewer than all frames in the group. The frames are indexed in a database in association with the generated context metadata. 1. A method comprising:analyzing a sequence of video frame to determine whether the sequence was captured by a static camera; andresponsive to determining that the sequence of frames was captured by the static camera, generating context metadata for each frame in the sequence based on an analysis of fewer than all frames in the sequence.2. The method of claim 1 , further comprising:determining that a select frame of the sequence was captured by a moving camera, andresponsive to the determination, subjecting the select frame to a first series of processing operations for generating the context metadata that is different than a second series of processing operations used to generate the context metadata for the frames in the sequence captured by the static camera.3. The method of claim 2 , wherein the first series of processing operations generates context metadata for the select frame based on an image analysis that is limited to the select frame.4. The method of claim 1 , wherein generating context metadata for each frame in the sequence based on the analysis of fewer than all frames in the sequence further comprises:selecting a keyframe from the sequence;generating at least a portion of the context metadata based on an analysis of the keyframe without analyzing other frames of the sequence; andindexing the other frames of the sequence in association with the generated context metadata.5. The method of claim 4 , wherein generating the context metadata further comprises:generating descriptors for multiple objects present in the keyframe.6. The method of claim 5 , wherein generating the context metadata further comprises:generating a ...

Подробнее

Номер записи: 6

03-01-2019 дата публикации

Method, apparatus and arrangement for summarizing and browsing video content

Номер: US20190005133A1

Автор: Chenzhi HE, Claire-Hélène Demarty, Gwenaelle Marquant, Matthieu Fradet

Принадлежит: THOMSON LICENSING

A method, apparatus and arrangement for the summarizing and browsing of video sequences are provided and include at least one horizontal strip having time-sequenced video frames belonging to a single video sequence and at least one vertical strip having a plurality of video frames belonging to different video sequences, each of the plurality of video frames of the at least one vertical strip having at least one feature in common. In one instance, the at least one horizontal strip and the at least one vertical strip are arranged to intersect at a video frame of the at least one horizontal video strip having the least one feature in common with the video frames in the at least one vertical strip.

Подробнее

Номер записи: 7

28-01-2021 дата публикации

RETRIEVAL DEVICE, TRAINING DEVICE, RETRIEVAL SYSTEM, AND RECORDING MEDIUM

Номер: US20210026887A1

Автор: FUKUSHIMA Shintaro, Sasai Takeyuki

Принадлежит: TOYOTA JIDOSHA KABUSHIKI KAISHA

The retrieval device extracts a feature corresponding to search text by inputting the search text into a pre-trained text feature extraction model. The retrieval device then, for plural combinations stored in a database associating a text description including plural sentences, with a vehicle-view video, and with vehicle behavior data representing temporal vehicle behavior, computes a text distance represented by a difference between a feature extracted from each sentence of the text description associated with the video and vehicle behavior data, and the feature corresponding to the search text. The retrieval device outputs as the search result a prescribed number of pairs of video and vehicle behavior data pairs in sequence from the smallest text distance according to the text distances. 1. A retrieval device , comprising:a memory, anda processor coupled to the memory, the processor being configured to:acquire a search text,extract a feature corresponding to the search text by inputting the search text to a text feature extraction model configured to extract features from input sentences, the text feature extraction model being pre-trained so as to reduce a loss represented by a difference between a feature extracted from a sentence and a feature extracted from a correctly matched vehicle-view video, and also being pre-trained so as to reduce a loss represented by a difference between a feature extracted from the sentence and a feature extracted from correctly matched vehicle behavior data representing temporal vehicle behavior,compute a text distance for each of a plurality of combinations stored in the memory, each combination associating a text description, including a plurality of sentences, with a vehicle-view video and with vehicle behavior data representing temporal vehicle behavior, the text distance being represented by a difference between a feature extracted from each sentence of the text description associated with the video and the vehicle behavior ...

Подробнее

Номер записи: 8

31-01-2019 дата публикации

Determining rate of video feed stories in a feed provided to users

Номер: US20190034438A1

Автор: Jie Xu, Zigang Xiao

Принадлежит: Facebook Inc

A social networking system provides feed to users in response to requests from users. The feed comprises feed stories that comprise one or more of text, image, or videos. The social networking system generates a feed in which the number of video feed stories is determined based on the video feed story rate determined for each specific request. The social networking system determines the video feed story rate based on various features including features describing the request, for example, the time of the day that the request is received. The social networking system may include other features for determining the video feed story rate, for example, user specific features describing the user that sent the request. The social networking system sends the generated feed to the client device of the target user.

Подробнее

Номер записи: 9

11-02-2016 дата публикации

Event-based image classification and scoring

Номер: US20160042249A1

Автор: Boris Babenko, Nils Peter Welinder

Принадлежит: Dropbox Inc

Embodiments of the present invention analyze and score each image associated with a group to determine representative image or images for the group. Such analysis can include detecting objects shown in the images, determining the quality of the images, and/or contextually analyzing the images as a group. In some embodiments, each image in a group (e.g., an event) of images can be analyzed by one or more image analysis modules that calculate a score for the image based on a different image characteristic. A composite image score can then be calculated based on the various image characteristic scores to identify the image or images to be used as to represent the group.

Подробнее

Номер записи: 10

09-02-2017 дата публикации

Hierarchical watermark detector

Номер: US20170039669A1

Автор: Adnan M. Alattar, Osama M. Alattar

Принадлежит: Digimarc Corp

The present invention relates generally to digital watermarking. One aspect of the disclosure includes a method comprising: obtaining data representing imagery; using one or more configured processors, analyzing a plurality of portions of the data to detect a watermark orientation component, said analyzing employing a match filter, in which the match filter yields a correlation value for each of the plurality of portions; determining a first portion from the plurality of portions that comprises a correlation value meeting a predetermined value; and directing a watermark decoder at the first portion to decode a plural-bit watermark payload, in which the watermark decoder produces a watermark signature for the first portion, and in which the watermark decoder searches a plurality of areas at or around the first portion to decode the plural-bit watermark payload. Of course, many other aspects and disclosure are provided in this patent document.

Подробнее

Номер записи: 11

13-02-2020 дата публикации

SYSTEMS AND METHODS TO TRANSFORM EVENTS AND/OR MOOD ASSOCIATED WITH PLAYING MEDIA INTO LIGHTING EFFECTS

Номер: US20200050074A1

Автор: Cremer Markus Kurt, Merchant Shashank, Vartakavi Aneesh

Принадлежит:

Example systems and methods to transform events and/or mood associated with playing media into lighting effects are disclosed herein. An example apparatus includes a content identifier to identify a first event occurring during presentation of media content at a first time. The example apparatus includes a content driven analyzer to determine a first lighting effect to be produced by a light-producing device based on the first event and instruct the light-producing device to produce the first lighting effect based on the first event during presentation of the media content. The content identifier is to identify a second media event occurring during presentation of the media content at a second time after the first time. The content driven analyzer is to instruct the light-producing device to one of maintain the first lighting effect based on the second event or produce a second lighting effect based on the second event during presentation of the media content. 1. A non-transitory machine readable storage medium comprising instructions that , when executed , cause at least one machine to at least:determine a color associated with media content during presentation of the media content;determine a light effect to be produced by a light-producing device based on the color; andinstruct the light-producing device to produce the light effect during presentation of the media content.2. The non-transitory machine readable storage medium of claim 1 , wherein the media content is video content and the instructions claim 1 , when executed claim 1 , cause the at least one machine to identify the color of one or more frames of the video content.3. The non-transitory machine readable storage medium of claim 1 , wherein the instructions claim 1 , when executed claim 1 , cause the machine to identify a dominant color in the one or more frames as the color.4. The non-transitory machine readable storage medium of claim 1 , wherein the lighting effect includes a first color of light to ...

Подробнее

Номер записи: 12

27-02-2020 дата публикации

SYSTEMS AND METHODS FOR VIDEO ARCHIVE AND DATA EXTRACTION

Номер: US20200065329A1

Автор: Abeloe Kenneth A., Hsu Dennis

Принадлежит:

Systems and methods for full motion video search are provided. In one aspect, a method includes receiving one or more search terms. The search terms include one or more of a characterization of the amount of man-made features in a video image and a characterization of the amount of natural features in the video image. The method further includes searching a full motion video database based on the one or more search terms. 1. A method of adding a video entity to a full motion video database supporting search capabilities , the method comprising:determining a starting cell in the full motion video database; and splitting the starting cell to create two new child cells, and', 'adding the video entity to one of the two new child cells., 'determining if the starting cell should be split, and if the starting cell should be split2. The method of claim 1 , wherein splitting the starting cell comprises splitting the starting cell along a split axis.3. The method of claim 2 , wherein the split axis is along a line of latitude.4. The method of claim 2 , wherein the split axis is along a line of longitude.5. The method of claim 2 , wherein the two new child cells are created along the split axis.6. The method of claim 1 , further comprising moving one or more video entities stored in the starting cell from the starting cell to the new child cells.7. The method of claim 6 , further comprising adding pointers to the starting cell pointing to the child cells.8. The method of claim 1 , wherein determining if the starting cell should be split comprises determining if the starting cell is full.9. The method of claim 8 , further comprising adding the video entity to the starting cell if the starting cell not full.10. The method of claim 1 , further comprising:determining a child cell from a plurality of child cells of the starting cell if the starting cell contains a plurality of child cells; andadding the video entity to one of the plurality of child cells.11. The method of claim 10 ...

Подробнее

Номер записи: 13

11-03-2021 дата публикации

DATA PROCESSING APPARATUS, DATA PROCESSING METHOD, AND NON-TRANSITORY STORAGE MEDIUM

Номер: US20210073515A1

Автор: Araki Takuya, HIRAKAWA Yasufumi, Liu Jianquan, Nishimura Shoji

Принадлежит: NEC Corporation

A data processing apparatus () of the present invention includes a unit that retrieves a predetermined subject from moving image data. The data processing apparatus includes a person extraction unit () that analyzes moving image data to be analyzed and extracts a person whose appearance frequency in the moving image data to be analyzed satisfies a predetermined condition among persons detected in the moving image data to be analyzed, and an output unit () that outputs information regarding the extracted person. 1. A data processing apparatus , comprising:at least one memory configured to store one or more instructions; and analyze data to be analyzed and extract a subject whose appearance frequency in the data to be analyzed satisfies a predetermined condition among subjects detected in the data to be analyzed; and', 'output information regarding the extracted subject,, 'at least one processor configured to execute the one or more instructions towherein the appearance frequency is the number of times each of the subjects appears in the data to be analyzed, or appearance rate of each of the subjects in the data to be analyzed.2. The data processing apparatus according to claim 1 , analyze the data to be analyzed in units of time windows, each time window having a predetermined time width,', 'determine whether each of the subjects appears in each of a plurality of the time windows, and', 'calculate the appearance frequency based on the number of time windows in which each of the subjects appears., 'wherein the processor is further configured to execute the one or more instructions to3. The data processing apparatus according to claim 1 , analyze moving image data to be analyzed and extract a person whose appearance frequency in the moving image data to be analyzed satisfies a predetermined condition among persons detected in the moving image data to be analyzed, and', 'output information regarding the extracted person., 'wherein the processor is further configured to ...

Подробнее

Номер записи: 14

18-03-2021 дата публикации

Media management system for video data processing and adaptation data generation

Номер: US20210081699A1

Автор: Avner Levi, Ika Bar-Menachem, Mor Geva PIPEK, Ohad Jassin, Olivier Nano, Ori ZIV, Oron NIR, Royi Ronen

Принадлежит: Microsoft Technology Licensing LLC

In various embodiments, methods and systems for implementing a media management system, for video data processing and adaptation data generation, are provided. At a high level, a video data processing engine relies on different types of video data properties and additional auxiliary data resources to perform video optical character recognition operations for recognizing characters in video data. In operation, video data is accessed to identify recognized characters. A video OCR operation to perform on the video data for character recognition is determined from video character processing and video auxiliary data processing. Video auxiliary data processing includes processing an auxiliary reference object; the auxiliary reference object is an indirect reference object that is a derived input element used as a factor in determining the recognized characters. The video data is processed based on the video OCR operation and based on processing the video data, at least one recognized character is communicated.

Подробнее

Номер записи: 15

05-03-2020 дата публикации

VALIDATING ASSET MOVEMENT USING VIRTUAL TRIPWIRES AND A RFID-ENABLED ASSET MANAGEMENT SYSTEM

Номер: US20200077054A1

Автор: MARIADOSS PANDIAN

Принадлежит:

Validation of asset movement can begin with the definition of a virtual tripwire for video cameras of a video surveillance system that monitor a defined area of interest through which RFID-identified assets are conveyed. The virtual tripwire can visually define a boundary within the view field of the video camera. The video camera can detect the movement of an asset through a virtual tripwire and an asset management system can be notified of the potential misdirection of the asset. The asset management system can obtain the asset's RFID data from a RFID handling system that monitors the area of interest. The validity of the detected movement can be determined using RFID data and an asset movement specification that defines pathing rules for assets and virtual tripwires. When the movement of the asset is determined to be invalid, a record of suspect movement data can be generated for the asset. 120-. (canceled)21. A method , implemented by an asset management system , for validating a movement of an asset having an attached radio frequency identification (RFID) tag , comprising: a virtual tripwire event associated with the asset,', 'RFID data from the attached RFID tag, and', 'an asset movement specification for the asset; and, 'determining that the movement of an asset is invalid based upon'}generating, based upon the determination that the movement is invalid, a suspect movement data record, wherein the RFID data,', 'information pertaining to the virtual trip event., 'the suspect movement data record includes22. The method of claim 21 , whereinthe information pertaining to the virtual trip event includes video data of the asset.23. The method of claim 22 , whereinthe video data includes video data from a video camera that captured the virtual tripwire event and video data, contemporaneous with the virtual tripwire event, from at least one other neighboring video camera.24. The method of claim 22 , whereinthe video data includes a predefined period of video relative ...

Подробнее

Номер записи: 16

21-04-2016 дата публикации

Method for obtaining a mega-frame image fingerprint for image fingerprint based content identification, method for identifying a video sequence, and corresponding device

Номер: US20160110609A1

Автор: Frédéric Lefebvre, Joaquin ZEPEDA SALVATIERRA, Patrick Perez

Принадлежит: Thomson Licensing SAS

A temporal section that is defined by boundary images is selected in a video sequence. A maximum of k stable image frames are selected in the temporal section of image frames having a lowest temporal activity. Image fingerprints are computed from the selected stable image frames. A mega-frame image fingerprint data structure is constructed from the computed fingerprints.

Подробнее

Номер записи: 17

09-04-2020 дата публикации

Importing State Data From A Video Stream Into A Gaming Session

Номер: US20200108310A1

Автор: Davis Paul, Doshi Nishant, Gabler Phillip, Grady Daniel T., Laubach Scott, Mundra Arvind

Принадлежит:

Methods, systems, and apparatuses are described for identifying computer programs/video games that are compatible with a particular content stream and obtaining metadata indicating various different states of the content stream or event depicted in the content stream. The metadata may be supplied as input state information to the computer programs/video games to begin a simulation of the event depicted in the content stream or to initiate a video game session corresponding to the state of the content stream. 1. A method comprising:determining a type of a content item being output;determining, based the type of the content item, a gaming content associated with the content item, wherein the gaming content is executable by a computing device;determining event state information indicating a current event state of the content item; andcausing, based on the event state information, a session of the gaming content to begin using the current event state of the content item.2. The method of claim 1 , wherein the current event state of the content item comprises a scenario in a live event and wherein the event state information comprises statistics associated with the scenario in the live event.3. The method of claim 1 , further comprising:receiving a user input indicating a request for the session of the gaming content to begin, wherein the determining the gaming content is based on the request.4. The method of claim 1 , further comprising:determining the gaming content associated with the content item is not in a list of a plurality of potential gaming contents associated with the content item; andsending a request to cause display of a request to obtain the gaming content.5. The method of claim 4 , further comprising:sending, based on a request to obtain the gaming content, a request to the computing device to download the gaming content from a gaming content repository; andadding the gaming content to the list of the plurality of potential gaming contents available to ...

Подробнее

Номер записи: 18

04-05-2017 дата публикации

System and method for multi-modal fusion based fault-tolerant video content recognition

Номер: US20170124096A1

Автор: Chun-Chieh Huang, Kuo-Don HSI, Yen-Cheng Chen

Принадлежит: VISCOVERY Pte Ltd

A system and a method for multi-modal fusion based fault tolerant video content recognition is disclosed. The method conducts multi-modal recognition on an input video to extract multiple components and their respective appearance time in the video. Next, the multiple components are categorized and recognized respectively via different algorithms. Next, when the recognition confidence of any component is insufficient, a cross-validation with other components is performed to increase the recognition confidence and improve the fault tolerance of the components. Furthermore, when the recognition confidence of an individual component is insufficient, the recognition continues and tracks the component, spatially and temporally when it applies, until frames of high recognition confidence in the continuous time period is reached. Finally, multi-modal fusion is performed to summarize and resolve any recognition discrepancies between the multiple components, and to generate indices for every time frame for the ease of future text-based queries.

Подробнее

Номер записи: 19

16-04-2020 дата публикации

METHODS, SYSTEMS, AND MEDIA FOR DETECTING TWO-DIMENSIONAL VIDEOS PLACED ON A SPHERE IN ABUSIVE SPHERICAL VIDEO CONTENT BY TILING THE SPHERE

Номер: US20200117908A1

Автор: Konrad Matthias, Pavetic Filip, Vorushin Roman

Принадлежит:

Methods, systems, and media for analyzing spherical video content are provided. More particularly, methods, systems, and media for detecting two-dimensional videos placed on a sphere in abusive spherical video content by tiling the sphere are provided. In some embodiments, the method comprises: receiving an identifier of a spherical video content item, wherein the spherical video content item has a plurality of views and wherein the spherical video content item is encoded into a plurality of two-dimensional video frames; selecting a first frame of the plurality of two-dimensional video frames associated with the spherical video content item; dividing the first frame into a plurality of tiles spanning the first frame of the spherical video content item; calculating, for each tile of the plurality of tiles, a probability that the tile includes content of a particular type of content; determining, for each tile, whether the probability exceeds a predetermined threshold; in response to determining, for a particular tile, that the probability exceeds the predetermined threshold, causing the content associated with the tile to be analyzed using a video fingerprinting technique; and in response to determining, using the video fingerprinting technique, that the content associated with the tile matches a reference content item of a plurality of reference content items, generating an indication of the match in association with the identifier of the spherical video content item. 1. A method for analyzing spherical video content , the method comprising:receiving, using a hardware processor, an identifier of a spherical video content item, wherein the spherical video content item has a plurality of views;selecting, using the hardware processor, a frame associated with the spherical video content item;projecting, using the hardware processor, the frame of the spherical video content item to a two-dimensional representation;dividing, using the hardware processor, the two- ...

Подробнее

Номер записи: 20

04-06-2015 дата публикации

Hierarchical watermark detector

Номер: US20150154724A1

Автор: Adnan M. Alattar, Osama M. Alattar

Принадлежит: Digimarc Corp

The present invention relates generally to digital watermarking. In one implementation, we provide a hierarchical digital watermark detector method. The method includes: i) in a first layer of a hierarchical search, performing watermark detection on blocks of at least a portion of an incoming suspect signal; ii) identifying a first block in the portion that is likely to include a decodable digital watermark; and iii) in a second layer of the hierarchical search, performing additional watermark detection on overlapping blocks in a neighborhood around the first block. Another implementation provides a hierarchical watermark detector including a buffer and a detector. The buffer stores portions of an incoming signal. The detector evaluates watermark detection criteria for blocks stored in the buffer, and hierarchically zooms into a neighborhood of blocks around a block associated with watermark detection criteria that satisfies detection criteria.

Подробнее

Номер записи: 21

15-09-2022 дата публикации

Image Search in Walkthrough Videos

Номер: US20220292135A1

Автор: Allen Thomas Friel, DeCamp Philip, Fleischman Michael Ben, Hein Gabriel

Принадлежит:

A spatial indexing system receives a set of walkthrough videos of an environment taken over a period of time and receives an image search query that includes an image of an object. The spatial indexing system searches the set of walkthrough videos for instances of the object. The spatial indexing system presents search results in a user interface, displaying in a first portion a 2D map associated with one walkthrough video with marked locations of instances of the object and a second portion with a histogram of instances of the object over time in the set of walkthrough videos. 1. A method comprising:receiving an image search query for a walkthrough video of an environment;searching the walkthrough video to identify portions of frames of the walkthrough video that correspond to the image search query;refining the image search query based on the identified portions of frames of the walkthrough video;searching a set of walkthrough videos of the environment using the refined image search query to identify additional portions of frames within the set of walkthrough videos that correspond to the refined image search query, each of the set of walkthrough videos associated with a timestamp; andgenerating an interface for display including a representation of a floorplan of the environment with marked locations within the representation of the floorplan corresponding to the identified portions of frames of the walkthrough video, and including a histogram, the independent axis of the histogram comprising a timeline corresponding to the timestamps of the set of walkthrough videos, the histogram including bars indicating instances of the identified additional portions of frames within the set of walkthrough videos.2. The method of claim 1 , wherein the received image search query comprises a selection of a portion of a frame of the walkthrough video.3. The method of claim 1 , wherein the received image search query comprises an image of an object or surface.4. The method of ...

Подробнее

Номер записи: 22

16-05-2019 дата публикации

EVENT SEARCH SYSTEM, DEVICE, AND METHOD

Номер: US20190146996A1

Автор: JANG Se Mi, LIM Jeong Eun, NOH Seung In

Принадлежит:

An event search device includes: a first communication unit configured to receive, from a camera, a plurality of image frames of a surveillance area generated in response to an occurrence of an event; and metadata including time information of the occurrence of the event and location information of an event in the image frames; a user interface configured to receive a search condition for searching for the event, the search condition including an event search area; a screen unit configured to display the event search area on the image frames; and a control unit configured to search for an event satisfying the search condition within the event search area by using the metadata, wherein at least one of a location, a size, and a shape of the event search area is defined through the user interface. 1. An event search system comprising:a sensor configured to detect an occurrence of an event and generate a signal;a camera in response to receiving the signal from the sensor, confiqured to generate a plurality of image frames of a surveillance area and metadata of the image frames, the metadata comprising time information of the occurrence of the event and location information of the event in the plurality of image frames; andan event search device configured to receive the plurality of image frames and the metadata from the camera and search for an event, the event search device comprising:a first communication unit configured to receive the plurality of image frames and the metadata,a user interface configured to receive for searching for the event, the search condition comprising an event search area;a screen unit configured to display the event search area on the image frames; anda control unit configured to search for an event satisfying the search condition within the event search area by using the metadata,wherein at least one of a location, a size, and a shape of the event search area is defined through the user interface.2. The event search system of claim 1 , ...

Подробнее

Номер записи: 23

15-09-2022 дата публикации

TRACKING MULTIPLE OBJECTS IN A VIDEO STREAM USING OCCLUSION-AWARE SINGLE-OBJECT TRACKING

Номер: US20220292828A1

Автор: CHAKRABORTY Ishani, Chen Yi-Ling, Yuan Lu

Принадлежит:

The disclosure herein enables tracking of multiple objects in a real-time video stream. For each individual frame received from the video stream, a frame type of the frame is determined. Based on the individual frame being an object detection frame type, a set of object proposals is detected in the individual frame, associations between the set of object proposals and a set of object tracks are assigned, and statuses of the set of object tracks are updated based on the assigned associations. Based on the individual frame being an object tracking frame type, single-object tracking is performed on the frame based on each object track of the set of object tracks and the set of object tracks is updated based on the performed single-object tracking. For each frame received, a real-time object location data stream is provided based on the set of object tracks. 1. A system for tracking multiple objects in a real-time video stream , the system comprising:at least one processor; determine a frame type of an individual frame of a plurality of frames of the real-time video stream as an object-detection frame type,', 'detect a set of object proposals in the individual frame, the set of object proposals including a first subset and a second subset,', 'associate the first subset of the set of object proposals with a set of active object tracks and the second subset of the set of object proposals with a set of passive object tracks,', 'update a status of the set of object tracks based on the association of the first subset of the set of object proposals and the second subset of the set of object proposals, and', 'process the individual frame in real-time, and, 'at least one memory comprising computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the at least one processor toan output device configured to provide an object location data stream based on the processing.2. The system of claim 1 , wherein claim ...

Подробнее

Номер записи: 24

17-06-2021 дата публикации

IMAGE PRE-PROCESSING METHOD, APPARATUS, AND COMPUTER PROGRAM

Номер: US20210182566A1

Автор: Jung Tae Young

Принадлежит:

The present invention relates to an image pre-processing method, apparatus, and computer program. The present invention relates to a method for processing an arbitrary image, comprising the steps of: dividing the image into scene units including one or more frames; selecting a frame to be searched according to a preset criterion from the scene; identifying an object associated with a preset subject from the frame to be searched; and searching for image corresponding to the object and/or object information and mapping the search result to the object. According to the present invention, the efficiency of an object-based image search can be maximized and resources to be used for image processing can be minimized. 1. A method for processing a video , the method comprising:dividing the video based on a scene comprising at least one frame;selecting a search target frame according to a preset criterion in the scene;identifying an object related to a preset subject in the search target frame; andsearching for at least one of an image or object information corresponding to the object and mapping search results to the object.2. The method as claimed in claim 1 , wherein the dividing of the video based on the scene comprises:identifying a color spectrum of the frame; anddistinguishing between scenes of a first frame and a second frame, which are consecutive, when a change in the color spectrum between the first frame and the second frame is greater than or equal to a preset threshold.3. The method as claimed in claim 1 , wherein the dividing of the video based on the scene comprises:detecting feature information estimated as an object in the frame;determining whether first feature information present in a first frame is present in a consecutive second frame; anddistinguishing between the scenes of the first frame and the second frame when the first feature information is not present in the second frame.4. The method as claimed in claim 1 , wherein the dividing of the video ...

Подробнее

Номер записи: 25

01-06-2017 дата публикации

Methods and systems for identifying an object in a video image

Номер: US20170154240A1

Автор: Alexander G. Ross, George Vogiatzis, Marco Fiocco, Noa Garcia

Принадлежит: Vloggsta Inc

A smart phone ( 101 ) captures an image ( 113 ) of a video frame of a movie and sends it to a server ( 107 ) for comparison with data stored in a database ( 112 ) for determining the title of the movie and identifying the particular video frame in the movie. Another database ( 110 ) is interrogated whereby items of merchandise contained in the image of the identified video frame are extracted and displayed on the smartphone along with brand names and a retailer's details. To reduce processing time and the amount of data stored, data relating to only a subset of the total video frames comprising a movie are stored in the database ( 112 ). The user may also request to have displayed on the smartphone's screen ( 102 ), one of several frames preceding or following the captured frame and request information on items of merchandise displayed in such frames.

Подробнее

Номер записи: 26

15-09-2022 дата публикации

METHOD, ELECTRONIC DEVICE, AND COMPUTER PROGRAM PRODUCT FOR DATA PROCESSING

Номер: US20220294867A1

Автор: He Bin, Hou Anzhou, Jia Zhen, Sha Danqing

Принадлежит:

Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for data processing. The method for data processing includes: generating, based on a category of a target data content segment, a target tag for the target data content segment; acquiring a reference tag set, a reference tag in the reference tag set being generated based on a category of a historical data content segment; and determining redundancy of the target data content segment based on comparison between the target tag and the reference tag set. In this way, the distribution of a large amount of redundant data from an edge server to a terminal device can be avoided, thereby saving bandwidth, improving delay, and further providing a better user experience for AR/VR content experience. 1. A method for data processing , comprising:generating, based on a category of a target data content segment, a target tag for the target data content segment;acquiring a reference tag set, a reference tag in the reference tag set being generated based on a category of a historical data content segment; anddetermining redundancy of the target data content segment based on comparison between the target tag and the reference tag set.2. The method according to claim 1 , wherein determining the redundancy of the target data content segment comprises:determining, based on the comparison between the target tag and the reference tag set, whether there is a historical data content segment matching the target data content segment in a historical data content segment set corresponding to the reference tag set; anddetermining, when determining that there is no matching historical data content segment, the target data content segment as a non-redundant data content segment.3. The method according to claim 2 , further comprising:sending the target data content segment to a terminal device.4. The method according to claim 2 , further comprising:adding the target tag to the reference ...

Подробнее

Номер записи: 27

22-09-2022 дата публикации

IMAGE DIFFERENCE DETECTION

Номер: US20220301283A1

Автор: JR. Elmer G., Musser

Принадлежит:

The described technology is generally directed towards comparing two images for content differences, such as images based on the frames of a show and a derivative version of that show. Frame pairs such as from an original show and its derivative version are processed into image pairs, which can include decoding, scaling, luminance extraction and/or filtering. An edge pixel image is obtained via edge detection for each image. Edge pixels in one image are compared against a counterpart edge pixel (and possibly neighboring pixels) in the other image to determine matches (matching edge pixels) and mismatches. An image with too many errors based on the mismatches is deemed as a candidate for further review. A difference image can be generated to assist a reviewer in detecting where the mismatches were detected. By repeating for the various frames, a show can be automatically compared against its derivative for content differences. 1. A system , comprising:a processor; and obtaining a source frame from a first group of frames;', 'obtaining a counterpart frame from a second group of frames;', 'processing the source frame into a source image comprising edge pixels and non-edge pixels;', 'processing the counterpart frame into a counterpart image comprising edge pixels and non-edge pixels;', 'determining mismatched edge pixels in the counterpart image that do not match corresponding edge pixels of the source image; and', 'reporting on the mismatched edge pixels., 'a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations, the operations comprising2. The system of claim 1 , wherein the determining the mismatched edge pixels in the counterpart image that do not match edge pixels of the source image comprises claim 1 , for an edge pixel of the source image claim 1 , searching a corresponding pixel in the counterpart image to look for an edge pixel.3. The system of claim 1 , wherein the determining the mismatched edge ...

Подробнее

Номер записи: 28

22-09-2022 дата публикации

METHOD AND APPARATUS FOR IDENTIFYING DISPLAY SCENE, DEVICE AND STORAGE MEDIUM

Номер: US20220301286A1

Автор: An Rongqiao, Chen Kehua, Chen Yuerong, Ni Zihan, Sun Yipeng, Yao Kun

Принадлежит:

A method and apparatus for identifying a display scene, a method and apparatus for training a model, a device, a storage medium and a computer program product are provided. An implementation of the method may comprise: acquiring a feature vector of a to-be-identified image and acquiring a base library feature vector set; ascertaining, from the base library feature vector set, at least two candidate feature vectors based on a similarity coefficient between the feature vector of the to-be-identified image and each feature vector in the base library feature vector set; performing threshold comparisons on similarity coefficients of the at least two candidate feature vectors, to obtain a target feature vector; and determining a display scene of the to-be-identified image based on a display scene tag corresponding to the target feature vector. 1. A method for identifying a display scene , comprising:acquiring a feature vector of a to-be-identified image and acquiring a base library feature vector set;ascertaining, from the base library feature vector set, at least two candidate feature vectors based on a similarity coefficient between the feature vector of the to-be-identified image and each feature vector in the base library feature vector set;performing threshold comparisons on similarity coefficients of the at least two candidate feature vectors, to obtain a target feature vector; anddetermining a display scene of the to-be-identified image based on a display scene tag corresponding to the target feature vector.2. The method according to claim 1 , wherein the at least two candidate feature vectors comprise one primary feature vector and at least one secondary feature vector claim 1 , and the performing the threshold comparisons on the similarity coefficients of the at least two candidate feature vectors to obtain the target feature vector comprises:in response to determining that a similarity coefficient of the primary feature vector is greater than a first threshold, ...

Подробнее

Номер записи: 29

22-09-2022 дата публикации

METHOD AND SYSTEM FOR SEMI-SUPERVISED CONTENT LOCALIZATION

Номер: US20220301308A1

Автор: Hsiao Jenhao

Принадлежит:

A special-purpose convolutional learning model architecture outputs a convolutional feature map at a last of its convolutional layers, then performs binary classification based on non-semantically labeled dataset. The convolutional feature map, containing a combination of low-spatial resolution features and high-spatial resolution features, in conjunction with a binary classification output of a special-purpose learning model having transferred learning from a pre-trained learning model, may be used to non-semantically derive a segmentation map. The segmentation map may reflect both low-spatial resolution and high-spatial resolution features of the original image on a one-to-one pixel correspondence, and thus may be utilized to highlight or obscure subject matter of the image in a contextually fitting manner at both a global scale and a local scale over the image, without semantic knowledge of the content of the image. 1. A method comprising:deriving a gradient of a feature vector of a convolutional feature map;deriving a feature map contribution parameter with regard to the feature vector from the gradient of the feature vector;obtaining a localization map by aggregating feature map contribution parameters; anddrawing an edge of a segmentation map based on values of the localization map.2. The method of claim 1 , wherein the gradient of the feature vector is derived with regard to a classification.3. The method of claim 2 , wherein deriving the gradient of the feature vector comprises computing a partial derivative of a probability score of a feature of the convolutional feature map with regard to the classification over a feature vector of the convolutional feature map.4. The method of claim 3 , wherein deriving the feature map contribution parameter comprises normalizing the partial derivative over each pixel of the convolutional feature map.5. The method of claim 4 , wherein normalizing the partial derivative comprises summing the partial derivative over each ...

Подробнее

Номер записи: 30

22-09-2022 дата публикации

INFORMATION PROCESSING APPARATUS, NONVERBAL INFORMATION CONVERSION SYSTEM, AND INFORMATION PROCESSING METHOD

Номер: US20220301347A1

Автор: TANAKA Satomi

Принадлежит: RICOH COMPANY, LTD.

An information processing apparatus includes circuitry. The circuitry estimates an intention indicated in first nonverbal information, which is nonverbal information of a first user, based on a feature amount of the first nonverbal information and a nonverbal expression model indicating a relation between the first nonverbal information and an intention. The circuitry converts the first nonverbal information into second nonverbal information to be output to a second user based on the estimated intention and a set conversion rule of the nonverbal information. 1. An information processing apparatus , comprising circuitry configured to:estimate an intention indicated in first nonverbal information, which is nonverbal information of a first user, based on a feature amount of the first nonverbal information and a nonverbal expression model indicating a relation between the first nonverbal information and an intention; andconvert the first nonverbal information into second nonverbal information to be output to a second user based on the estimated intention and a set conversion rule of the nonverbal information.2. The information processing apparatus of claim 1 , wherein the circuitry is further configured toacquire video data obtained by capturing the first user,output conversion data obtained by converting the acquired video data,estimate the intention based on the first nonverbal information indicated in the video data,convert the video data indicating the first nonverbal information into the conversion data indicating the second nonverbal information, andoutput a video represented by the conversion data to a reception apparatus used by the second user.3. The information processing apparatus of claim 2 , whereinthe nonverbal expression model is a model that learns individuality of the first user in nonverbal expression, andthe circuitry calculates an estimated value of the intention, the estimated value being obtained by inputting the acquired video data to the ...

Подробнее

Номер записи: 31

30-05-2019 дата публикации

Image display apparatus and method of operating the same

Номер: US20190163702A1

Автор: Ernesto Evgeniy SANCHES SHAYDA, Ja-Yoon Koo, Jatin GARG, Vivek Agarwal

Принадлежит: SAMSUNG ELECTRONICS CO LTD

An image display apparatus for automatically determining an image in a video having a best picture of an object selected by a user.

Подробнее

Номер записи: 32

21-05-2020 дата публикации

METHOD AND APPARATUS FOR PROCESSING REQUESTS FOR THUMBNAIL IMAGES IN A VIDEO SURVEILLANCE SYSTEM

Номер: US20200159768A1

Автор: HENRIKSEN Nicolai

Принадлежит:

In a video surveillance system comprising a plurality of recording servers, each storing video data received from at least one video surveillance camera, a method of processing requests for thumbnail images comprises receiving a thumbnail request to a main queue . The thumbnail request includes information indicating a specific video camera and a specific recording server. The thumbnail request is allocated to a recording server queue corresponding to the specific recording server. It is determined whether an idle resource processor associated with the recording server queue is available to process the thumbnail request. If no idle resource processor is available, then it is determined how many resource processors are associated with the recording server queue and, if this is less than a maximum number, a new resource processor is initiated for processing the thumbnail request and the new resource processor is put into an idle state after processing the thumbnail request. 1. A method of processing requests for thumbnail images in a video surveillance system comprising a plurality of recording servers , each storing video data received from at least one video surveillance camera , comprising:receiving a thumbnail request to a main queue, the thumbnail request including information indicating a specific video camera from which a thumbnail image is requested and a specific recording server on which video data from the specific video camera is stored;allocating the thumbnail request to a recording server queue corresponding to the specific recording server;determining whether an idle resource processor associated with the recording server queue is available to process the thumbnail request;if an idle resource processor is available, processing the thumbnail request using the resource processor; (i) determining how many resource processors are associated with the recording server queue;', '(ii) if this is less than a maximum number, initiating a new resource processor ...

Подробнее

Номер записи: 33

21-06-2018 дата публикации

Methods, systems, and media for detecting two-dimensional videos placed on a sphere in abusive spherical video content by tiling the sphere

Номер: US20180173957A1

Автор: Filip Pavetic, Matthias Konrad, Roman Vorushin

Принадлежит: Google LLC

Methods, systems, and media for analyzing spherical video content are provided. More particularly, methods, systems, and media for detecting two-dimensional videos placed on a sphere in abusive spherical video content by tiling the sphere are provided. In some embodiments, the method comprises: receiving an identifier of a spherical video content item, wherein the spherical video content item has a plurality of views and wherein the spherical video content item is encoded into a plurality of two-dimensional video frames; selecting a first frame of the plurality of two-dimensional video frames associated with the spherical video content item; dividing the first frame into a plurality of tiles spanning the first frame of the spherical video content item; calculating, for each tile of the plurality of tiles, a probability that the tile includes content of a particular type of content; determining, for each tile, whether the probability exceeds a predetermined threshold; in response to determining, for a particular tile, that the probability exceeds the predetermined threshold, causing the content associated with the tile to be analyzed using a video fingerprinting technique; and in response to determining, using the video fingerprinting technique, that the content associated with the tile matches a reference content item of a plurality of reference content items, generating an indication of the match in association with the identifier of the spherical video content item.

Подробнее

Номер записи: 34

28-05-2020 дата публикации

SYSTEMS AND METHODS FOR FLEXIBLY USING TRENDING TOPICS AS PARAMETERS FOR RECOMMENDING MEDIA ASSETS THAT ARE RELATED TO A VIEWED MEDIA ASSET

Номер: US20200167386A1

Автор: AGARWAL Vineet, VENKATARAMAN Sashikumar

Принадлежит:

Systems and methods are provided herein for flexibly using trending topics as parameters for recommending media assets that are related to a viewed media asset. A media guidance application may determine that a user has viewed a media asset. The media guidance may identify a plurality of attributes corresponding to the viewed media asset and determine that a respective attribute of the plurality of attributes matches a trending topic. The media guidance application may update a set of weightings corresponding to the plurality of attributes by increasing a weighting corresponding to the respective attribute and adjust a recommendation for a media asset different from the viewed media asset based on the updated set of weightings. The media guidance application may generate for display the recommendation of the media asset different from the viewed media asset. 1. A method comprising:determining that a user has viewed a media asset;identifying a plurality of attributes corresponding to the viewed media asset;determining that a respective attribute of the plurality of attributes matches a trending topic;updating a set of weightings corresponding to the plurality of attributes by increasing a weighting corresponding to the respective attribute;adjusting a recommendation for a media asset different from the viewed media asset based on the updated set of weightings; andgenerating for display the recommendation of the media asset different from the viewed media asset.2. A method for flexibly using trending topics as parameters for recommending media assets that are related to a viewed media asset , the method comprising:determining, based on processing input signals at user equipment, that a user has viewed a media asset;identifying a plurality of attributes corresponding to the viewed media asset;comparing each of the plurality of attributes corresponding to the viewed media asset to a plurality of topics with which a threshold number of people have interacted within a ...

Подробнее

Номер записи: 35

18-09-2014 дата публикации

Videolens media engine

Номер: US20140270507A1

Автор: Priyan Gunatilake

Принадлежит: Sony Corp

A system, method, and computer program product for automatically analyzing multimedia data are disclosed. Embodiments receive multimedia data, detect portions having specified features, and output a corresponding subset of the multimedia data. Content features from downloaded or streaming movies or video clips are identified as a human probably would do, but in essentially real time. Embodiments then generate an index or menu based on individual consumer preferences. Consumers can peruse the index, or produce customized trailers, or edit and tag content with metadata as desired. The tool can categorize and cluster content by feature, to assemble a library of scenes or scene clusters according to user-selected criteria.

Подробнее

Номер записи: 36

11-06-2020 дата публикации

PROVIDING RELEVANT COVER FRAME IN RESPONSE TO A VIDEO SEARCH QUERY

Номер: US20200183977A1

Автор: Dhawan Anmol, Duggal Ashish, Soni Sachin

Принадлежит:

The present disclosure is directed towards methods and systems for providing relevant video scenes in response to a video search query. The systems and methods identify a plurality of key frames of a media object and detect one or more content features represented in the plurality of key frames. Based on the one or more detect content features, the systems and methods associate tags indicating the detected content features with the plurality of key frames of the media object. The systems and methods, in response to receiving a search query including search terms, compare the search terms with the tags of the selected key frames, identify a selected key frame that depicts at least one content feature related to the search terms, and provide a preview image of the media item depicting the at least one content feature. 1. A method of identifying an image to represent relevant content features within a digital video , the method comprising:analyzing, by at least one processor, a media object comprising a plurality of frames of video content to determine a plurality of key frames within the video content;identifying, by the at least one processor, a content feature depicted in each key frame of the plurality of key frames;associating, by the at least one processor, a tag with each key frame of the plurality of key frames, wherein a given tag corresponds to a given content feature depicted in a given key frame of the plurality of key frames;determining to provide the media object as a search result to a search query received from a client device;selecting a key frame from the plurality of key frames as a preview image for the media object based on the search query relating to a particular content feature depicted in the selected key frame as indicated by a particular tag associated with the selected key frame; andproviding, to the client device, the preview image for the media object in response to the search query.2. The method of claim 1 , wherein:the search query ...

Подробнее

Номер записи: 37

10-08-2017 дата публикации

Content-Based Video Representation

Номер: US20170228599A1

Автор: Paloma de Juan

Принадлежит: Yahoo Holdings Inc

Methods and systems for classifying a video include analyzing an image captured in each frame of the video file to identify one or more elements. Each element identified in the image of each frame is matched to a corresponding term defined in a vocabulary list. A number of frames within the video file in which each of the element that correspond to the term in the vocabulary list, appears, is determined. A vector is generated for the video file identifying each term in the vocabulary list. The vector is represented as a name-value pair with the name corresponding to the term in the vocabulary list and the value corresponding to number of frames in which the element matching the term appears in the video file.

Подробнее

Номер записи: 38

19-08-2021 дата публикации

Methods, systems, and media for generating search results based on contextual information

Номер: US20210256055A1

Автор: Ant Oztaskent, Ingrid McAulay Trollope, Yaroslav Volovich

Принадлежит: Google LLC

Methods, systems, and media for generating search results based on contextual information are provided. In some implementations, a method for presenting search results is provided, the method comprising: receiving, using a hardware processor, a query related to media that is currently being presented; identifying a program that is currently being presented; identifying a plurality of keywords associated with the identified program; determining one or more of the plurality of keywords that are contextually relevant to the query; obtaining a plurality of search results based on the query and the one or more contextually relevant keywords; and causing at least one of the plurality of search results to be presented to the user.

Подробнее

Номер записи: 39

19-08-2021 дата публикации

GRAPH CONVOLUTIONAL NETWORKS FOR VIDEO GROUNDING

Номер: US20210256059A1

Автор: DAS SUBHRO, Gan Chuang, Liu Sijia, WANG Dakuo, Zhang Yang

Принадлежит:

Method and apparatus that includes receiving a query describing an aspect in a video, the video including a plurality of frames, identifying multiple proposals that potentially correspond to the query where each of the proposals includes a subset of the plurality of frames, ranking the proposals using a graph convolution network that identifies relationships between the proposals, and selecting, based on the ranking, one of the proposals as a video segment that correlates to the query. 1. A method comprising:receiving a query describing an aspect in a video, the video comprising a plurality of frames;identifying multiple proposals that potentially correspond to the query, wherein each of the proposals comprises a subset of the plurality of frames;ranking the proposals using a graph convolution network (GCN) that identifies relationships between the proposals; andselecting, based on the ranking, one of the proposals as a video segment that correlates to the query.2. The method of claim 1 , wherein ranking the proposals comprises:identifying visual features in the proposals using a visual feature encoder; andgenerating query features from the query using a recurrent neural network (RNN).3. The method of claim 2 , wherein ranking the proposals comprises:generating a graph comprising nodes and edges based on the visual features and the query features; andinputting the graph into the GCN for processing.4. The method of claim 3 , wherein ranking the proposals comprises:updating node features for the nodes in the graph; andcalculating edge weights for the edges in the graph.5. The method of claim 3 , wherein ranking the proposals comprises:performing node aggregation; andranking the proposals based on the node aggregation and results from processing the graph using the GCN.6. The method of claim 1 , wherein at least two of the proposals comprise overlapping frames of the plurality of frames in the video.7. The method of claim 6 , wherein at least two of the proposals ...

Подробнее

Номер записи: 40

19-08-2021 дата публикации

SYSTEMS AND METHODS FOR SEARCHING FOR EVENTS WITHIN VIDEO CONTENT

Номер: US20210256060A1

Автор: Janakiraman Kirupakar, Muthusamy Baskaran, Patel Keyurbhai

Принадлежит:

A video management system (VMS) may search for one or more events in a plurality of video streams captured and stored at a plurality of remote sites. The VMS may generate time-stamped metadata for each video stream captured at the remote site. The time-stamped metadata for each video stream may identify one or more objects and/or events occurring in the corresponding video stream as well as an identifier that uniquely identifies the corresponding video stream. Each of the plurality of remote sites may send the time-stamped metadata to a central hub, wherein the time-stamped metadata may be stored in a data lake, and a user may enter a query into a video query engine, wherein the video query engine may be operatively coupled to the central hub. 1. A method for searching for one or more events in a plurality of video streams captured and stored at a plurality of remote sites , the method comprising:generating, at each of the plurality of remote sites, time-stamped metadata for each video stream captured at the remote site, the time-stamped metadata for each video stream identifying one or more objects and/or events occurring in the corresponding video stream as well as an identifier that uniquely identifies the corresponding video stream;sending, by each of the plurality of remote sites, the time-stamped metadata to a central hub, wherein the time-stamped metadata is stored in a data lake;entering a query by a user into a video query engine, wherein the video query engine is operatively coupled to the central hub;applying the query to the time-stamped metadata stored in the data lake to search for one or more objects and/or events in the plurality of video streams that match the query;returning a search result to the user, wherein the search result identifies one or more matching objects and/or events in the plurality of video streams that match the query, and for each matching object and/or event that matches the query, providing a link to the corresponding video ...

Подробнее

Номер записи: 41

27-08-2015 дата публикации

Logo presence detection based on blending characteristics

Номер: US20150242698A1

Автор: Faisal Ishtiaq, Kevin L. Baum

Принадлежит: Arris Technology Inc

A video processing system detects an overlay image, such as a logo, in a picture of a video stream, the overlay for example being a broadcaster's logo. The detection is based on evaluation of blending characteristics of a picture frame. The method of detection of an overlay defines first and second areas within the image, the first and second areas being non-overlapping. Next an alpha-blended value is calculated for the mean color value of the second area with an overlay color value. Then, if the mean color value of the first area is closer to the alpha-blended value than it is to the mean color value of the second area, the overlay can be indicated as detected and defined within the picture. Detection of the overlay can be used to identify an owner of the video, or detect when a scene change such as a commercial occurs.

Подробнее

Номер записи: 42

25-07-2019 дата публикации

OBJECT SEARCH BY DESCRIPTION

Номер: US20190228040A1

Автор: Gur Ran, MILMAN Michael, NAOR Nir, Ostrovsky-Berman Yaron, Wronsky Ran

Принадлежит: VERINT SYSTEMS LTD.

Systems and methods search video data for objects that satisfy a general object description. A database is populated with identified objects and object characteristics detected in video data with at least one identifier that specifies video image data. At least one search parameter is received that presents a general object description. The database is queried based upon the received at least one search parameter. At least one identifier is returned from the database based upon the at least one search parameter. 110-. (canceled)11. A method of Image data storage and retrieval , the method comprising:acquiring video image data from a plurality of cameras, the acquired video image data comprising an identifier indicative of a camera of the plurality of cameras that acquired the video image data and a time stamp of when the video image data was acquired;detecting a plurality of objects in the acquired video image data from the plurality of cameras;detecting, for each detected object, object characteristics in the acquired video image data;obtaining, for each detected object, other object characteristics from a source other than the acquired video image data;aggregating, for each detected object, (i) the object characteristics detected in the acquired video image data and (ii) at least a portion of the object characteristics obtained from a source other than the acquired video image data;storing, in a database, each detected object, each detected object's aggregated characteristics and each detected object's associated identifier;receiving search parameters and Boolean operators at a searcher engine operating on a processor, wherein the search parameters correspond to a user's estimate of an objects characteristics;querying the database for object characteristics based upon the received search parameters and Boolean operators;returning identifiers meeting the query based upon the received search parameters and Boolean operators;computing a similarity score for each ...

Подробнее

Номер записи: 43

16-07-2020 дата публикации

MOUNTED DISPLAYS THAT AUTOROTATE TO MATCH CONTENT ON DISPLAY

Номер: US20200225696A1

Автор: Chapman Steven M., Deuel Matthew, Patel Mehul, POPP JOSEPH

Принадлежит: Disney Enterprises, Inc.

Some implementations of the disclosure are directed to automatically rotating displays to display media content based on metadata extracted from the media content that provides an indication of a target display orientation to display the media content. In one implementation, a method includes: detecting media content for display on a display, wherein the display is mounted on a rotatable display mount; extracting metadata from the detected media content, the extracted metadata providing an indication of a target display orientation to display the media content; using at least the extracted metadata, automatically causing the rotatable display mount to rotate the display to the target orientation; and displaying the media content on the rotated display. 1. A method , comprising:detecting content for display on a display, wherein the display is mounted on a rotatable display mount;extracting metadata from the content, the metadata providing an indication of a target display orientation;rotating the rotatable display mount to rotate the display toward the target display orientation based on the metadata;displaying the content on the rotated display; andresponsive to detecting an obstruction to rotation of the rotatable display mount to the target display orientation, displaying a second content on the display, wherein the second content is different than the content.2. The method of claim 1 , wherein the metadata includes an aspect ratio or resolution of the content claim 1 , wherein the aspect ratio or the resolution is used to determine the target display orientation.3. The method of claim 1 , further comprising during the rotation of the rotatable display mount claim 1 , using a sensor to determine a presence of the obstruction.4. The method of claim 3 ,wherein the second content includes an audible alert or a visual alert.5. The method of claim 1 , further comprising:detecting a third content for display on the display,extracting an additional metadata from the ...

Подробнее

Номер записи: 44

26-08-2021 дата публикации

SYSTEMS AND METHODS FOR CONTROLLING DISPLAY OF VIDEO CONTENT IN AN ONLINE MEDIA PLATFORM

Номер: US20210263966A1

Автор: BECK Jeffrey Lee, CLARKIN Phillip P., RAMPERSAD Ravi, WOJCICKI Jeffrey John

Принадлежит: The Toronto-Dominion Bank

A processor-implemented method is disclosed. The method includes: obtaining user account data identifying one or more tradeable objects associated with an entity and a resource allocation profile in connection with the one or more tradeable objects; receiving, from at least one media data source, video data for one or more videos, the video data including content indicator tags associated with the one or more videos; identifying a first subset of the one or more videos based on: comparing the content indicator tags associated with the one or more videos with identifying information associated with the one or more tradeable objects; and selecting videos for inclusion in the first subset based on the comparing and the resource allocation profile; and sending, to the client device, an indication of the first subset of videos for display in a user interface on the client device. 1. A computing system , comprising:a communications module communicable with an external network;a memory; and obtain user account data identifying one or more tradeable objects associated with an entity and a resource allocation profile in connection with the one or more tradeable objects;', 'receive, from at least one media data source, video data for one or more videos, the video data including content indicator tags associated with the one or more videos;', performing textual comparison between the content indicator tags associated with the one or more videos and identifiers associated with the one or more tradeable objects; and', 'selecting videos for inclusion in the first subset based on the comparing and filtering using preferences of the entity defined in the resource allocation profile; and, 'identify a first subset of the one or more videos based on, 'send, to a client device, an indication of the first subset of videos for display in a user interface on the client device., 'a processor coupled to the communications module and the memory, the processor being configured to2. The ...

Подробнее

Номер записи: 45

28-10-2021 дата публикации

System, Device, and Method for Generating and Utilizing Content-Aware Metadata

Номер: US20210334547A1

Автор: Cohen-Tidhar Amnon, Lev-Ami Tal

Принадлежит:

System, device, and method for generating and utilizing content-aware metadata, particularly for playback of video and other content items. A method includes: receiving a video file, and receiving content-aware metadata about visual objects that are depicted in said video file; and dynamically adjusting or modifying playback of that video file, on a video playback device, based on the content-aware metadata. The modifications include content-aware cropping, summarizing, watermarking, overlaying of other content elements, modifying playback speed, adding user-selectable indicators or areas around or near visual objects to cause a pre-defined action upon user selection, or other adjustments or modification. Optionally, a modified and content-aware version of the video file is automatically generated or stored. Optionally, the content-aware metadata is stored internally or integrally within the video file, in its header or as a private channel; or is stored in an accompanying file. 1. A method comprising:(a) receiving a video file, and receiving content-aware metadata about visual objects that are depicted by said video file;(b) dynamically adjusting playback of said video file, on a video playback device, based cumulatively on at least (I) said content-aware metadata, and (II) dimensions of a screen-portion that is currently available for playback of said video file on said video playback device.2. The method of claim 1 ,wherein step (a) comprises: receiving a video file which comprises at least (i) a video channel carrying video data, and (ii) a private channel carrying content-aware metadata; wherein both the video channel carrying the video data and the private channel carrying the content-aware metadata are internal components of said video file;wherein step (b) comprises: extracting said content-aware metadata from said private channel of said video file, and dynamically adjusting the playback of the video based on content-aware metadata that was extracted from ...

Подробнее

Номер записи: 46

22-09-2016 дата публикации

Method and apparatus for automatically sending multimedia file, mobile terminal, and storage medium

Номер: US20160275077A1

Автор: Weixin Zhang

Принадлежит: ZTE Corp

A method for automatically sending a multimedia file comprises: obtaining a voice feature of each object in a multimedia file ( 101 ); matching the obtained voice feature of each object with a voice feature of each contact in a voice parameter database ( 102 ); and when matching succeeds, automatically sending the multimedia file to a contact obtained by successful matching ( 103 ). Further disclosed are an apparatus for automatically sending a multimedia file, a mobile terminal, and a storage medium.

Подробнее

Номер записи: 47

18-11-2021 дата публикации

METHOD FOR OBTAINING ONLINE PICTURE-BOOK CONTENT AND SMART SCREEN DEVICE

Номер: US20210357452A1

Автор: DAI Xiaocheng, LANG Liu, Wang Le

Принадлежит:

The present disclosure provides a method for obtaining online picture-book content, and relates to a field of online education technologies. The method is applied to the smart screen device. The method includes: establishing a communication link between the smart screen device and a prism component; identifying an offline picture book by using the prism component and obtaining image content of the offline picture book; and obtaining online picture-book content in a video form based on the image content.

Подробнее

Номер записи: 48

20-08-2020 дата публикации

SEARCHING RECORDED VIDEO

Номер: US20200265085A1

Автор: Aghdasi Farzin, Millar Greg, Wang Lei

Принадлежит:

Embodiments of the disclosure provide for systems and methods for creating metadata associated with a video data. The metadata can include data about objects viewed within a video scene and/or events that occur within the video scene. Some embodiments allow users to search for specific objects and/or events by searching the recorded metadata. In some embodiments, metadata is created by receiving a video frame and developing a background model for the video frame. Foreground object(s) can then be identified in the video frame using the background model. Once these objects are identified they can be classified and/or an event associated with the foreground object may be detected. The event and the classification of the foreground object can then be recorded as metadata. 1. A method comprising:receiving a search query from a user through a user interface, wherein the search query includes information for searching for either or both a classification of an object and an event associated with an object;retrieving metadata files associated with the search query, wherein the metadata files comprise information regarding either or both object classifications and object events within a video frame;searching the retrieved metadata files for metadata that matches the search query; andproviding a listing of video segments that match the search query through the user interface.2. The method according to claim 1 , wherein the search query comprises either or both of object classifications and object events and the searches for metadata that match either or both of object classifications and object events.3. The method according to further comprising:receiving an indication from the user identifying a video segment in the listing of video segments;retrieving the indicated video segment; anddisplaying the retrieved video segment to the user.4. The method according to claim 1 , wherein the search query comprises information identifying an object location in the field of view of a ...

Подробнее

Номер записи: 49

27-09-2018 дата публикации

Hierarchical watermark detector

Номер: US20180276782A1

Автор: Adnan M. Alattar, Osama M. Alattar

Принадлежит: Digimarc Corp

The present invention relates generally to digital watermarking. One aspect of the disclosure includes a method comprising: obtaining data representing imagery; using one or more configured processors, analyzing a plurality of portions of the data to detect a redundantly embedded watermark signal, the analyzing producing detection statistics for each of the plurality of portions, the detection statistics comprising a payload signature, a rotation angle and a scale factor for each portion of the plurality of portions; accumulating payload signatures based on compatible rotation angles and scale factors, said accumulating yielding an accumulated payload signature; and decoding a plural-bit payload from the accumulated payload signature. Of course, many other aspects and disclosure are provided in this patent document.

Подробнее

Номер записи: 50

13-10-2016 дата публикации

Systems and methods for defining and analyzing video clusters based on video image frames

Номер: US20160300110A1

Автор: Ladislav Prosek

Принадлежит: Facebook Inc

Systems, methods, and non-transitory computer-readable media can identify a first video represented based on a first set of image frames. A first subset of image frames can be extracted from the first set of image frames. The first subset of image frames can be compared to one or more image frames associated with a collection of video clusters. It can be determined that at least a threshold quantity of image frames in the first subset matches, within an allowable deviation, at least some image frames associated with a first video cluster included the collection of video clusters. The first video cluster can be defined to include the first video.

Подробнее

Номер записи: 51

20-10-2016 дата публикации

Process for generating a video tag cloud representing objects appearing in a video content

Номер: US20160307044A1

Автор: Abdelkader Outtagarts, Corinne Obled, Emmanuel Marilly, Fabien Diot, Sylvain Squedin

Принадлежит: Alcatel Lucent SAS

Process for generating a video tag cloud representing objects appearing in a video content, said process providing: a step (B) for extracting video frames of said video content and individually segmenting said video frames into regions; a step (C) for building, for each extracted frame, a topology graph for modelizing the space relationships between the segmented regions of said frame; a step (D) for extracting from the set of built topology graphs frequent patterns according to spatial and temporal constraints, each pattern comprising at least one segmented region; a step (E) for regrouping frequent patterns representing parts of a same object by using trajectories constraints, so as to detect frequent objects of said video content; a step (F) for determining, for each detected frequent object, a weighting factor to apply to said object according at least to spatial and temporal constraints used for extracting the patterns of said object and to trajectories constraints used to regroup said patterns; a step (H) for generating a video tag cloud comprising a visual representation for each of said frequent objects according to their weighting factors.

Подробнее

Номер записи: 52

03-11-2016 дата публикации

Systems and methods for analyzing time series data based on event transitions

Номер: US20160321257A1

Автор: Jixu Chen, MING-CHING Chang, Peter Henry Tu, Siwei Lyu, Yelin Kim

Принадлежит: Morpho Detection LLC

A method for analyzing time series data to identify an event of interest is provided. The method includes receiving, at a computing device, time series data that includes the event of interest, identifying, using the computing device, a start time of the event of interest and an end time of the event of interest by modeling at least one transitional pattern in the time series data, and categorizing, using the computing device, the event of interest based on the at least one transitional pattern.

Подробнее

Номер записи: 53

03-10-2019 дата публикации

Systems and methods for determining video content relevance

Номер: US20190303499A1

Автор: Marc Sharma

Принадлежит: CBS INTERACTIVE INC

Systems and methods for making video content recommendations. Metadata relating to at least one content item consumed by the user is received. The video data is stored as at least one video data file for each of the content items and frame change times are extracted for each of the content items from corresponding of the at least one video data file. Frame image files are created for each of the content items based on corresponding sets of the frame times and entity data is extracted from the frame files. Audio data of each of the content items is converted to text data and the entity data and the text data are merged for each content item to create a list of tokens corresponding to each content item. A document vector is determined for each content item based on the list of tokens corresponding to that content item and the similarity of each item of content to each item in a different set of content items is determined based on the vectors. Recommendations of content are presented to a user based on the scores.

Подробнее

Номер записи: 54

01-10-2020 дата публикации

ALIGNING CONTENT PLAYBACK WITH VEHICLE TRAVEL

Номер: US20200309547A1

Автор: DAVIDSSON Staffan, Tollesson Vincent

Принадлежит:

Techniques regarding aligning content playback with a vehicle's estimated travel time are provided. For example, one or more embodiments described herein can comprise a system, which can further comprise a processor that executes computer executable components stored in memory. The system can also comprise a travel component that can receive information regarding vehicle navigation. Further, the system can comprise a ranking component that can rank relevancy of classified content based upon preferences and context of an entity in a vehicle. Moreover, the system can comprise a content selection component that can select a first set of content for play to the entity based in part on estimated travel time of the vehicle and the relevancy of the classified content. 1. A system , comprising:a processor that executes computer executable components stored in memory;a travel component that receives information regarding vehicle navigation;a ranking component that ranks relevancy of classified content based upon preferences and context of an entity in a vehicle; anda content selection component that selects a first set of content for play to the entity based in part on estimated travel time of the vehicle and the relevancy of the classified content.2. The system of claim 1 , further comprising an editing component that edits the first set of content based on the estimated travel time of the vehicle.3. The system of claim 2 , further comprising a prioritization component that prioritizes subsets of the first set of content as a function of relevancy to or preference of the entity claim 2 , wherein the editing component prunes the subsets of the first set of content as a function of the relevancy or preference of the entity.4. The system of claim 1 , further comprising a routing component that re-routes the vehicle based on play of the classified content.5. The system of claim 2 , further comprising a presentation component that controls presentation of the pruned content ...

Подробнее

Номер записи: 55

08-10-2020 дата публикации

METHOD AND APPARATUS FOR GROUNDING A TARGET VIDEO CLIP IN A VIDEO

Номер: US20200320303A1

Автор: He Dongliang, HUANG Jizhou, Li Fu, Liu Xiao, Wen Shilei, Zhao Xiang

Принадлежит:

A method and an apparatus for grounding a target video clip in a video are provided. The method includes: determining a current video clip in the video based on a current position; acquiring descriptive information indicative of a pre-generated target video clip descriptive feature, and executing a target video clip determining step which includes: determining current state information of the current video clip, wherein the current state information includes information indicative of a feature of the current video clip; generating a current action policy based on the descriptive information and the current state information, the current action policy being indicative of a position change of the current video clip in the video; the method further comprises: in response to reaching a preset condition, using a video clip resulting from executing the current action policy on the current video clip as the target video clip. 1. A method for grounding a target video clip in a video , comprising:determining a current video clip in the video based on a current position; determining current state information of the current video clip, wherein the current state information includes information indicative of a feature of the current video clip;', 'generating a current action policy based on the descriptive information and the current state information, the current action policy being indicative of a position change of the current video clip in the video; and, 'acquiring descriptive information indicative of a pre-generated target video clip descriptive feature, and executing a target video clip determining step which includesthe method further comprising:in response to reaching a preset condition, using a video clip resulting from executing the current action policy on the current video clip as the target video clip.2. The method according to claim 1 , further comprising:in response to failing to reach a preset condition, using a video clip resulting from executing the current ...

Подробнее

Номер записи: 56

29-10-2020 дата публикации

HASH-BASED MEDIA SEARCH

Номер: US20200342023A1

Автор: Diggins Jonathan, Phillips Gerard

Принадлежит:

A method of identifying an item of video content involves providing a spatial hash value and a temporal hash value for each image in a video collection. Each hash value is based on a measure of the entropy in differences between pixel values. A table of the pair of hash values against timecode is created and ordered according to one of the hash values. A search for a given pair of hash values can then be confined to that part of the table that matches the first value. 1. A system for identifying an item of video content , the apparatus comprising: a spatial search hash value relating to difference values between pixel values within each search image in the respective query sequence, and', 'a temporal search hash value relating to difference values between a pixel value in a first search image in the respective query sequence and a corresponding pixel value in a second search image that is temporally separated within the respective query sequence;, 'a query sequence processor configured to define a plurality of query sequences by a traveling window of search images of a plurality of temporally separated images of the video content, and to form a pair of search hash values for each of the plurality of query sequences, wherein the pair of search hash values comprisesa relational database configured to store a look up table that defines ordered search data corresponding to a first hash value of the pair of search hash values for each query sequence; anda search processor configured to search the ordered search data that corresponds to a sought hash value of the first hash value of the pair of search hash values to locate a sought hash value of a second hash value of the respective pair of hash values.2. The system according to claim 1 , wherein the query sequence processor is configured to execute instructions stored on a memory to define the plurality of query sequences and to determine the pair of search hash values for each of the plurality of query sequences.3. The ...

Подробнее

Номер записи: 57

14-12-2017 дата публикации

Using audio and video matching to determine age of content

Номер: US20170357654A1

Автор: Johan Georg Granström, Matthias Rochus Konrad, Oleg Bochkarev

Принадлежит: Google LLC

A computer at a content management system receives a first digital content item from a content provider. The computer matches the first digital content item to each of a plurality of reference digital content items in a database. The system determines a plurality of match metrics from the matches. Each match metric is indicative of a similarity between the first digital content item and one of the plurality of reference digital content items. Responsive to one of the match metrics being greater than a threshold level, the system sets a content age of the first digital content item to equal a content age of a reference digital content item associated with the match metric. Responsive to none of the match metrics being greater than the threshold, the system sets the content age of the first digital content item to a time of receiving the first digital content item.

Подробнее

Номер записи: 58

21-11-2019 дата публикации

VIDEO INGESTION FRAMEWORK FOR VISUAL SEARCH PLATFORM

Номер: US20190354766A1

Автор: MOORE Jimmy Daniel, MOORE Stephen Maurice, MURRAY Larry Patrick

Принадлежит:

This invention includes a framework and method to enable the transformation of video content into a format that allows for mapping and, hence, identification of individual sections (scene segments or otherwise) of the media content. It includes a means of ingesting images and video frames into a database. Consumer products can be matched with the images and/or objects in the images. A consumer can access the content by submitting a digital image taken on a mobile device. One or more objects in the user submitted image can be matched with products and/or commercial/promotional materials. 111.-. (canceled)12. A method of compiling frames from video , saving and indexing the frames in a database comprising the steps of:a) obtaining video;b) analyzing the video for features;c) separating the video into segments based on features, wherein segments are identified by grouping frames that correspond to the same or similar scenes, wherein duplicate segments in the video are removed, wherein segments in the video already present in the database are removed, wherein individual frames in the segments are extracted and visual features are generated for each frame and wherein the segments are annotated with metadata;d) analyzing visual features of segments to group segments that share visual features;e) analyzing suitability of each frame;f) providing an alert if unsuitability is determined;g) annotating one or more frames of segments with metadata if suitability is determined, wherein the visual features is extracted into a feature vector, wherein each feature vector is combined with metadata and unique identifiers to create a composite data structure comprising a frame identifier, segment identifier, frame visual features and metadata; andh) saving the one or more frames of segments into the database, wherein the database is queried using the composite data structure to determine if linked content exists in the database and either i) creates a segment insertion record and ...

Подробнее

Номер записи: 59

03-12-2020 дата публикации

IMAGE PROCESSING METHOD, STORAGE MEDIUM, AND COMPUTER DEVICE

Номер: US20200380031A1

Автор: CHENG Pei, Fu Bin, GAO Yu, ZHOU Jingjin

Принадлежит:

An image processing method includes: obtaining audio data corresponding to a reality scene image acquired in real time; dynamically determining attribute information of a virtual object according to the audio data, the attribute information indicating a visual state of the virtual object; identifying a target object from the reality scene image; determining, according to the target object, a fusion location of the virtual object determined according to the attribute information in the reality scene image according to the target object; fusing the virtual object determined according to the attribute information into the reality scene image according to the fusion location, the virtual object presenting different visual states that correspond to different attribute information dynamically determined according to the audio data. 1. An image processing method , applied to an image processing device , comprising:obtaining audio data corresponding to a reality scene image acquired in real time;dynamically determining attribute information of a virtual object according to the audio data, the attribute information indicating a visual state of the virtual object;identifying a target object from the reality scene image;determining, according to the target object, a fusion location of the virtual object in the reality scene image; andfusing the virtual object determined according to the attribute information into the reality scene image according to the fusion location, the virtual object presenting different visual states that correspond to different attribute information dynamically determined according to the audio data.2. The method according to claim 1 , wherein the obtaining audio data corresponding to a reality scene image acquired in real time comprises at least one of:acquiring the audio data corresponding to the reality scene image from a current environment in real time when acquiring the reality scene image in real time; orreading the audio data corresponding to a ...

Подробнее

Номер записи: 60

12-12-2019 дата публикации

RGB-D Camera Based Tracking System and Method Thereof

Номер: US20190377952A1

Автор: BABU Benzun Pious Wisely, Kim Soohwan, Ren Liu, Yan Zhixin

Принадлежит: ROBERT BOSCH GMBH

A visual SLAM system comprises a plurality of keyframes including a keyframe, a current keyframe, and a previous keyframe, a dual dense visual odometry configured to provide a pairwise transformation estimate between two of the plurality of keyframes, a frame generator configured to create keyframe graph, a loop constraint evaluator adds a constraint to the receiving keyframe graph, and a graph optimizer configured to produce a map with trajectory. 1. A method for computing visual Simultaneous localization and Mapping (SLAM) comprising:generating, by a visual odometry module, a local odometry estimate;generating, by a keyframe generator, keyframes;creating keyframe graph;adding constraint to the keyframe graph using a loop constraint evaluator; andoptimizing the keyframe graph with trajectory.2. The method of further comprising:generating a new keyframe between a keyframe and a current frame before generating a local odometry estimate.3. The method of wherein adding constraint to the keyframe graph using a loop constraint evaluator is based on a loop closure;wherein the loop closure is the return to previously visited locations.4. The method of claim 3 , further comprising adjusting a pose graph based on edge heights of different constraints in the keyframe graph after optimization.5. A method of applying a probabilistic sensor model for a dense visual odometry comprising:generating, by a keyframe generator, keyframes;creating keyframe graph;adding constraint to the keyframe graph using a loop constraint evaluator; andoptimizing the keyframe graph with trajectory6. The method of further comprising:generating a new keyframe between a keyframe and a current frame before generating a local odometry estimate.7. The method of wherein adding constraint to the keyframe graph using a loop constraint evaluator is based on a loop closure;wherein the loop closure is the return to previously visited locations.8. The method of claim 7 , further comprising adjusting a pose graph ...

Подробнее

Номер записи: 61

12-12-2019 дата публикации

METHOD AND APPARATUS FOR PROCESSING VIDEO

Номер: US20190377956A1

Автор: LUO Yu, Zheng Yehan

Принадлежит:

Embodiments of the present disclosure disclose a method and apparatus for processing a video. A specific embodiment of the method comprises: acquiring a target video and target video element information of the target video; extracting, based on the target video element information, a target video clip from the target video; obtaining, based on a preset corresponding relationship between video element information and a keyword determining method for a video clip, a keyword representing a category of the target video clip; and matching the keyword and with preset tag information set to obtain tag information of the target video clip, and associating and storing the target video clip and the tag information. 1. A method for processing a video , comprising:acquiring a target video and target video element information of the target video;extracting, based on the target video element information, a target video clip from the target video;obtaining, based on a preset corresponding relationship between video element information and a keyword determining method for a video clip, a keyword representing a category of the target video clip; andmatching the keyword with a preset tag information set to obtain tag information of the target video clip, and associating and storing the target video clip and the tag information.2. The method according to claim 1 , wherein the target video element information is an image frame included in the target video; and the extracting claim 1 , based on the target video element information claim 1 , a target video clip from the target video comprises:performing, for the image frame included in the target video, image recognition on the image frame to obtain an image recognition result for the image frame; anddetermining a video clip comprising image frames having associated image recognition results and arranged consecutively in the target video as the target video clip;or,the target video element information is a set of text information ...

Подробнее

Номер записи: 62

10-12-2020 дата публикации

SYSTEMS AND METHODS FOR AUTOMATIC GENERATION OF BOOKMARKS UTILIZING FOCUSED CONTENT ANALYSIS

Номер: US20200387540A1

Автор: CHEN Hung-En

Принадлежит:

A computing device initiates a video desktop recording capturing activities on a desktop displayed on the computing device, wherein during the video desktop recording, the computing device is further configured to detect at least one window within the video desktop recording, and identify a target window from among the at least one window detected within the video desktop recording. The computing device is further configured to generate a bookmark comprising metadata associated with the target window responsive to identifying the target window, the metadata comprising a time stamp when the target window was identified. The computing device is further configured to store the bookmark as part of the video desktop recording. 1. A method implemented in a computing device , comprising: detecting at least one window within the video desktop recording;', 'identifying a target window from among the at least one window detected within the video desktop recording;', 'responsive to identifying the target window, generating a bookmark comprising metadata associated with the target window, the metadata comprising a time stamp when the target window was identified; and', 'storing the bookmark as part of the video desktop recording., 'initiating a video desktop recording capturing activities on a desktop displayed on the computing device, wherein during the video desktop recording, performing the steps of2. The method of claim 1 , wherein the metadata further comprises: a title of the target window; a process identifier associated with the target window assigned by an operating system executing on the computing device claim 1 , and a descriptor relating to content within the target window.3. The method of claim 1 , wherein the metadata further comprises a graphical representation of the target window.4. The method of claim 1 , wherein a new bookmark is generated responsive to a new target window being detected.5. The method of claim 1 , wherein identifying the target window from ...

Подробнее

Номер записи: 63

17-12-2020 дата публикации

ELECTRONIC DEVICE AND METHOD FOR PROVIDING SERVICE INFORMATION RELATED TO BROADCAST CONTENT IN ELECTRONIC DEVICE

Номер: US20200394389A1

Автор: CHOI Gippeum, KIM Jongho, LEE Jungkun

Принадлежит:

An electronic device according to one embodiment of the present invention comprises: at least one communication interface; a display; a memory; and at least one processor electrically connected to the at least one communication interface, the display, and the memory, wherein the memory may store instructions which, when executed, cause the at least one processor to: acquire, in response to receiving a request for service information related to broadcast content sent from a content server, at least one image frame included in the broadcast content; determine whether at least one face region has been detected within the at least one image frame; perform, if the at least one face region is determined to have been detected, image analysis on the basis of the detected at least one face region; and provide, through the display, service information corresponding to a result of the image analysis. In addition, various embodiments are possible. 1. An electronic device , comprising:at least one communication interface;a display;a memory; andat least one processor electrically connected with the at least one communication interface, the display, and the memory, whereinthe memory stores instructions that cause, when executed, the at least one processor to:in response to reception of a request for service information related to broadcast content transmitted from a content server, obtain at least one image frame included in the broadcast content,determine whether at least one facial area is detected from the at least one image frame,upon determining that the at least one facial area is detected, perform image analysis based on the detected at least one facial area, andprovide service information corresponding to a result of the image analysis via the display.2. The electronic device of claim 1 , wherein the memory stores instructions that cause claim 1 , when executed claim 1 , the at least one processor to:set an object extraction area based on the detected at least one facial ...

Подробнее

Номер записи: 64

17-12-2020 дата публикации

METHOD AND APPARATUS FOR TRAINING FEATURE EXTRACTION MODEL, COMPUTER DEVICE, AND COMPUTER-READABLE STORAGE MEDIUM

Номер: US20200394416A1

Автор: GONG Guoping, Wu Tao, XU Xuyuan

Принадлежит: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED

Aspects of the disclosure provide a method and an apparatus for training a feature extraction model, a computer device, and a computer-readable storage medium that belong to the field of video processing technologies. The method can include detecting a plurality of images in one or more sample videos and obtaining at least two images including the same object. The method can further include determining the at least two images including the same object as sample images, and training according to the determined sample images to obtain the feature extraction model, where the feature extraction model is used for extracting a video feature of the video. 1. A method for training a feature extraction model that is performed by a computer device , the method comprising:detecting a plurality of images in one or more sample videos and obtaining at least two images including a same object, the at least two images indicating a change of the object in a time dimension;determining the at least two images including the same object as sample images; andtraining based on the determined sample images to obtain the feature extraction model that is configured to extracting a video feature of video.2. The method according to claim 1 , wherein the detecting the plurality of images in the one or more sample videos further comprises:detecting a first image in the one or more sample videos to determine a first object included in the first image; andtracking the first object in an image other than the first image to obtain a second image including the first object.3. The method according to claim 2 , wherein the tracking further comprises:tracking the first object in the image other than the first image to obtain a plurality of other images including the first object; and 'an image with a distance from the first image greater than a preset distance, where the distance between two images is a quantity of images between the two images, or a last image from the plurality of other images.', ' ...

Подробнее

Номер записи: 65

22-12-2022 дата публикации

RECORDING APPARATUS, METHOD OF CONTROLLING SAME, EXTERNAL APPARATUS, AND STORAGE MEDIUM

Номер: US20220405325A1

Автор: Komiyama Tomoaki, Minoshima Toshio

Принадлежит:

A recording apparatus to record, to a recording medium, a moving image file including video data and an additional information file, which is a file of additional information of the video data; and a control unit configured to perform control to record identification information for identifying category information of the additional information file or a data structure of the additional information file on the recording medium as a file other than the additional information file. 1. A recording apparatus comprising:at least one processor configured to function as: a recording unit configured to record, to a recording medium, a moving image file including video data and an additional information file, which is a file of additional information of the video data; anda control unit configured to perform control to record identification information for identifying category information of the additional information file or a data structure of the additional information file on the recording medium as a file other than the additional information file.2. The recording apparatus according to claim 1 , whereinthe control unit performs control to record the category information of the additional information file in the moving image file including the video data.3. The recording apparatus according to claim 1 , whereinthe control unit performs control to record the category information of the additional information file on a file other than the moving image file including the video data.4. The recording apparatus according to claim 3 , whereinthe control unit records the category information of the additional information file on a management information file for recording management information of the video data.5. The recording apparatus according to claim 1 , whereinthe control unit performs control to record the additional information file on the recording medium as an Extensible Markup Language (XML) file.6. The recording apparatus according to claim 5 , whereinthe control ...

Подробнее

Номер записи: 66

22-12-2022 дата публикации

TRAINING DATA GENERATION FOR ADVANCED FREQUENCY MANAGEMENT

Номер: US20220406038A1

Автор: AlDarabsah Khaldun Matter Ahmad, Assiotis Marios, Gallagher Joseph, Geng Hailong, KAWALE Jaya, Mazaheri Amir, Rotblat Mark Alden, She Chang, Tanaka Yoshihiro, WANG Haofei, Zhao Yu Tao, Zhong Chiyu

Принадлежит:

Systems and methods for programmatic generation of training data, including: a training data generation engine configured to: identify an image asset corresponding to an entity; identify a training video; select a consecutive subset of frames of the training video based on a procedure for ranking frames on their candidacy for overlaying content; for at least one frame of the subset of frames: perform an augmentation technique on the identified logo image to generate an augmented image asset; overlay at least one variation of the image asset, including the augmented image asset, onto each of the subset of frames to generate a set of overlayed frames; and generate an augmented version of the training video including the overlayed frames; and a model training engine configured to: train an artificial intelligence model for entity detection using the augmented version of the training video. 1. A system for programmatic generation of training data , comprising:a computer processor; and identify an image asset corresponding to an entity;', 'identify a training video;', 'select a consecutive subset of frames of the training video based on a procedure for ranking frames on their candidacy for overlaying content;', 'for at least one frame of the subset of frames: perform an augmentation technique on the identified logo image to generate an augmented image asset;', 'overlay at least one variation of the image asset, including the augmented image asset, onto each of the subset of frames to generate a set of overlayed frames; and', 'generate an augmented version of the training video comprising the overlayed frames; and, 'a training data generation engine executing on the computer processor and configured to enable the computer processor to 'train an artificial intelligence model for entity detection using the augmented version of the training video.', 'a model training engine configured to2. The system of claim 1 , wherein the procedure for ranking frames on their candidacy ...

Подробнее

Номер записи: 67

22-12-2022 дата публикации

SYSTEMS AND METHODS FOR THE ANALYSIS OF MOVING OBJECTS

Номер: US20220406058A1

Автор: BAUER TREVOR, BAUER WARREN

Принадлежит:

Methods and apparatus for the analysis of moving objects. In one embodiment, a pitch tracking system is disclosed that includes one or more cameras that have been positioned in a desired location and a computing system containing executable software that is configured to receive imaging data from the one or more cameras that captured the desired location and analyze seam placement and/or finger placement as a function of time for the received imaging data. Computer readable media and methodologies are also disclosed herein. 1. (canceled)2. A computer-implemented method for determining finger placement with respect to a pitched baseball , comprising the steps of:capturing a plurality of frames of a scene of the pitched baseball;segmenting an object of interest out of the captured scene by performing a thresholding operation on a pixel-by-pixel basis to determine a location of the object of interest within a first frame of the plurality of frames and utilizing the determined location of the object of interest within the first frame to assist in determining a location of the object of interest within a second frame of the plurality of frames;analyzing the segmented scene to identify markers associated with the object of interest; andcomparing the analyzed scene with a prior captured scene to enable a comparison in performance for the pitched baseball.3. The computer-implemented method of claim 2 , wherein the performing of the thresholding operation on the pixel-by-pixel basis comprises starting the thresholding operation in an area of the captured scene dependent upon a handedness for a pitcher of the pitched baseball.4. The computer-implemented method of claim 2 , wherein the performing of the thresholding operation on the pixel-by-pixel basis for the second frame of the plurality of frames is only performed on a subset of pixels of the second frame.5. The computer-implemented method of claim 4 , wherein the determining of the location of the object of interest ...

Подробнее

Номер записи: 68

14-01-2020 дата публикации

基于长时增强特征增强及稀疏动态采样的快速行为检测方法

Номер: CN110688918A

Автор: 徐树公, 李昱希, 林巍峣

Принадлежит: Shanghai Jiaotong University, UNIVERSITY OF SHANGHAI FOR SCIENCE AND TECHNOLOGY

一种基于长时增强特征增强及稀疏动态采样的快速行为检测方法，对视频直接进行时空联合的三维特征提取后进行时域检测，利用全局的长时信息对三维特征进行长时增强，根据时域检测结果对长时增强的特征沿时间维度在时域区间内进行稀疏动态采样，得到空间检测结果后与时域检测结果相结合，利用每个特征向量及其偏移向量计算检测框的相似度并通过逐帧聚类的方法实现检测框的链接，得到的检测框的集合，即行为管道。本发明能够显著提升分类精度以及定位精度从而最终有效提升检测的准确率，同时能够提升视频行为检测的运行效率。

Подробнее

Номер записи: 69

11-05-2016 дата публикации

一种车辆信息识别方法及系统

Номер: CN105574485A

Автор: 余扬, 莫尚林, 蒋君伟, 陈学健

Принадлежит: Beijing Sixiang Network Information Technology Co Ltd

本发明提供一种车辆信息识别方法及系统，方法包括：在全景图像的四个角点处设定角点矩形区域，对获取的目标全景图像进行扫描，得到多个初始矩形区域；根据预先设定的四个角点矩形区域，从扫描得到的多个初始矩形区域中筛选出至少一个候选矩形区域；对至少一个候选矩形区域分别进行分类分析，得到车辆信息所对应区域的目标图像；对目标图像进行字符分割，以及对分割后的字符进行识别。根据本方案，筛选出的候选矩形区域的个数远远小于初始矩形区域的个数，只需将数量较少的候选矩形区域进行分类分析即可，省却了大量分类分析时间，从而可以提高车辆信息所对应区域的图像的确定时间，进而提高车辆信息识别的效率。

Подробнее

Номер записи: 70

22-02-2017 дата публикации

一种同时定位与地图构建方法

Номер: CN106446815A

Автор: 刘浩敏, 章国锋, 鲍虎军

Принадлежит: Zhejiang University ZJU

本发明公开了一种同时定位与地图构建方法，该方法能够可靠地处理强烈旋转和快速运动。该方法提出了一种基于关键帧的同时定位与地图构建算法框架，能够支持快速的局部地图扩张。在此框架下，提出一种新的基于多个单应矩阵的特征跟踪方法，该方法在强烈旋转和快速运动下高效而鲁棒。又提出了一种基于滑动窗口的相机方位优化框架，用模拟的或实际的IMU数据对连续帧之间增加运动约束。最后，提出了一种获得特定平面和场景真实尺度的方法，从而把虚拟物体以真实大小摆放于特定平面之上。

Подробнее

Номер записи: 71

28-09-2018 дата публикации

一种基于深度语义分割海洋溢油检测系统与方法

Номер: CN108596065A

Автор: 李岩, 杨小飞

Принадлежит: Shenzhen Polytechnic

本发明提供了一种基于深度语义分割海洋溢油检测系统，包括计算服务器、GPU集群、海洋污染遥感影像数据库、正常无污染图像数据库，所述计算服务器部署在GPU集群上，所述计算服务器存储了训练好的神经网络模型，进行输入图像的识别。本发明还提供了一种基于深度语义分割海洋溢油检测方法。本发明的有益效果是：可以实现快速准确监测海洋溢油污染。

Подробнее

Номер записи: 72

23-04-2021 дата публикации

一种基于图像处理技术的廊桥工作状态检测系统及方法

Номер: CN110210427B

Автор: 刘星俞, 杨杰, 王旭辉, 许伟村, 许玉斌, 黄荣顺

Принадлежит: China Academy of Civil Aviation Science and Technology

本发明公开了一种基于图像处理技术的廊桥工作状态检测系统及方法，包括视频采集系统、数据采集系统、卷积神经网络系统和廊桥工作状态检测系统，所述视频采集系统、数据采集系统、卷积神经网络系统、廊桥工作状态检测系统依次通信连接，所述数据采集系统包括数据图像帧电子线划分模块和数据图像帧采集模块，所述卷积神经网络系统内部具有廊桥位置样本数据库，所述廊桥工作状态检测系统包括廊桥工作状态判断模块；所述视频采集系统包括若干个视频摄像头。本发明通过对机坪监控视频信号自动分析，判断廊桥当前的工作情况，可以同时得到廊桥靠机的开始和完成时间，以及廊桥离机的开始和完成时间。

Подробнее

Номер записи: 73

15-09-2020 дата публикации

视频匹配方法、装置、计算机设备和存储介质

Номер: CN111666922A

Автор: 周康明, 戚风亮

Принадлежит: Shanghai Eye Control Technology Co Ltd

本申请涉及一种视频匹配方法、装置、计算机设备和存储介质。包括：将获取的视频源输入至特征提取网络中进行特征提取，得到视频源中各个视频帧对应的特征图和特征向量；视频源包括至少一个视频帧，每个视频帧中均包括待检测对象；将各视频帧对应的特征图输入至相似度提取网络中进行相似度计算，得到视频源中各相邻两个视频帧之间的相似度；基于各相邻两个视频帧之间的相似度，对各个视频帧对应的特征向量进行融合处理，得到视频源对应的融合特征；根据视频源对应的融合特征和预设的视频库，在视频库中确定与视频源匹配的目标视频序列；目标视频序列包括至少两个目标视频源。采用本方法能够提高视频匹配结果的准确性。

Подробнее

Номер записи: 74

08-06-2021 дата публикации

视频分类方法、视频处理方法、装置、移动终端及介质

Номер: CN110348367B

Автор: 郭冠军

Принадлежит: Beijing ByteDance Network Technology Co Ltd

本公开实施例公开了一种视频分类方法、视频处理方法、装置、移动终端及介质。包括：根据视频中第一组图像帧的场景为预设场景的概率确定视频的第一分类标签，以及将视频中第二组图像帧输入图像序列分类模型，得到视频的第二分类标签；如果第一分类标签与第二分类标签一致，则得到视频的分类结果；如果第一分类标签与第二分类标签不一致，则将第二组图像帧重新输入其他图像序列分类模型，根据其他图像序列分类模型的输出确定视频的分类结果。本公开实施例可以根据采用不同的分类方式确定的多个视频分类结果确定视频的视频分类结果，可以在视频分类结果不一致时，根据多个图像序列分类模型的输出重新确定视频分类结果，可以准确地对视频进行分类。

Подробнее

Номер записи: 75

11-10-2019 дата публикации

基于视频监控的人群统计方法及系统

Номер: CN107133607B

Автор: 张亚妮, 黄良军

Принадлежит: Shanghai Institute of Technology

本发明提出一种基于视频监控的人群统计方法及系统，该方法包括以下步骤：S1：获取预监控区域的实时监控图像；S2：对实时监控图像进行前景分割，分割出人群；S3：将完整的人体特征模型分为五个局部特征模型，并为各局部特征模型配置相应的权重；S4：对分割后的人群的特征进行提取，并与所述五个局部特征模型进行匹配检测，根据权重及提取的五个局部特征模型与完整的人体特征模型之间的位置偏移计算综合匹配程度，若超过特定阈值，则匹配成功；S5：根据匹配成功的次数统计人群数量，绘制人群数量变化曲线，当曲线或曲线趋势触发异常事件时，进行警示。本发明的基于视频监控的人群统计方法及系统，可适用于不同场合，准确度较高。

Подробнее

Номер записи: 76

16-09-2015 дата публикации

Method for searching for film and television video

Номер: CN104915433A

Автор: 张永平, 李俊, 王宇

Принадлежит: NINGBO UNIVERSITY OF TECHNOLOGY

一种影视视频检索方法，涉及一种利用台本与字幕检索视频片段的方法，具有视频检测提取、对白信息提取标记、台本镜头搜索关键字获取、对话信息对白信息匹配、视频镜头与台本镜头匹配、视频镜头搜索关键字获取等步骤。与现有技术相比，本发明的有益效果是：通过视频的对白信息和完成台本的对话信息让视频镜头与台本镜头相对应，在检索时只要检索到台本镜头就能关联到视频镜头，既可以直接通过文字检索出视频片段，解决了以往视频片段无法直接检索，需要通过观看视频寻找相应视频片段的问题。

Подробнее

Номер записи: 77

22-02-2019 дата публикации

Crop straw burning monitoring method based on monitor video, device

Номер: CN106339657B

Автор: 张�杰, 李士进

Принадлежит: Individual

本发明公开了一种基于监控视频的秸秆焚烧监测方法。本发明方法基于实时性好且易于获取的监控视频，利用视频图像处理技术以及模式识别技术对视频图像进行分析处理，对视频监控区域内的秸秆焚烧事件进行智能识别并报警。本发明方法首先训练分类器，并利用下采样预处理方法克服正负样本不平衡问题，然后利用分类器对秸秆焚烧事件进行初步识别，再利用反映烟火图像梯度边缘信息的颜色梯度幅度值对分类器的初步识别结果进行修正，从而可得到更准确的监测结果。本发明还公开了一种基于监控视频的秸秆焚烧监测装置。相比现有技术，本发明可利用监控视频对秸秆焚烧进行全智能监测，且监测结果准确度更高，实时性更好。

Подробнее

Номер записи: 78

04-10-2016 дата публикации

Activity recommendation based on a context-based electronic files search

Номер: US9460200B2

Автор: James R. Kraemer, Josko Silobrcic, Robert R. Friedlander

Принадлежит: International Business Machines Corp

A computer hardware-implemented method, system, and/or computer program product generate an activity recommendation based on an identified non-synthetic event element in an electronic file. A context-based search of a database of electronic files is performed to identify a common synthetic event in all of the electronic files, where the synthetic event is a non-executable descriptor of a set of context-related factors. Each electronic file that contains the common synthetic event is a relevant electronic file. Each of the relevant electronic files are searched for at least one non-synthetic event element, wherein the non-synthetic event element is an event element that is absent from the synthetic event. A recommendation, based on the identified non-synthetic event element, is generated. A set of binary data, which includes the synthetic event, the relevant electronic file, the non-synthetic event element, and the recommendation, is then transmitted to a client.

Подробнее

Номер записи: 79

02-02-2018 дата публикации

A kind of visual experience Enhancement Method based on instant computing and dynamic tracing

Номер: CN103955499B

Автор: 夏天晗, 张恿, 李鲲, 杨明, 王皓悦

Принадлежит: BEIHANG UNIVERSITY

本发明公开了一种基于即时计算与动态追踪的视觉体验增强方法，根据图像边缘提取、轮廓匹配、模式匹配、SIFI特征检测与比较来进行图像的多场景适应性匹配，识别图像，反馈图片对应的动态信息，在移动端播放，同时根据图像主体判断与识别来锁定图像，跟踪主体的位置，通过动态信息播放，动态信息元件移动修正，使动态信息正确的覆盖图像，达到动态信息替换静态信息的目的，实现动态的视觉体验。该方法以动态信息“替换”静态信息，使动态信息“融入现实”从而实现视觉体验增强的效果，搭配各种移动终端载体，可广泛应用于平面媒体，博物馆，教育，旨在大幅度提升静态图片所携带的信息量，丰富图像表现形式，增强动态的视觉体验。

Подробнее

Номер записи: 80

08-05-2018 дата публикации

Image recognition monitoring method, system, computing device and readable storage medium storing program for executing

Номер: CN108009497A

Автор: 陈年春

Принадлежит: ZTE ICT Technologies Co Ltd

本发明提出了一种图像识别监控方法、系统、一种计算机设备以及可读存储介质。其中图像识别监控方法包括：获取待判断图像；将待判断图像与识别模型进行比较，识别待判断图像是否包含物体特征，得到识别结果；发送识别结果。通过本方案环保部门可以实时获取煤场的现场图像，并通过识别模型得到煤场的违规情况，无需委派专人对煤场进行监控，避免了工作人员重复往返拍摄图像，杜绝了煤场利用拍摄间隔钻空子，保证了煤场的防尘网的有效覆盖，从而减少煤场粉尘对于空气的污染。

Подробнее

Номер записи: 81

06-04-2016 дата публикации

Video condensation method

Номер: CN105469425A

Автор: 阳海华

Принадлежит: Shanghai Junesh Information Technology Co Ltd

本发明公开了一种视频浓缩方法，通过混合高斯背景模型检测运动目标的前景，统计每一帧图像上的像素点，建立背景，并使用统计差分对每个像素点进行分类；对运动物体通过搜索连通区域，将那些被分割开的区域进行连通区域合并；对运动目标进行跟踪，利用Kalman滤波器结合最小空间距离法来预测每个运动目标的坐标位置，并进行更新、跟踪、清除运动消失的运动对象，并记录运动目标的质心坐标、速度并统计；对运动轨迹进行优化，实现将不同时间段出现的运动目标平移到同一时间段，将前景粘贴到背景，并且不出现目标丢失和严重重叠。本发明实现监控视频的快速浏览，提高视频数据的利用率。

Подробнее

Номер записи: 82

29-10-2019 дата публикации

Video analysis method, apparatus, server and storage medium

Номер: CN110390262A

Автор: 盖超

Принадлежит: Ping An Technology Shenzhen Co Ltd

一种视频分析方法，包括：接收摄像头采集的视频图像；检测所述视频图像中的目标对象得到所述目标对象的类别；跟踪所述视频图像中的目标对象得到所述目标对象的状态；根据所述目标对象的类别和所述目标对象的状态分析得到所述视频图像中包含的业务场景；判断所述业务场景是否出现异常；及当所述视频图像中的业务场景出现异常时，记录所述业务场景出现异常时的关键信息。本发明还提供一种视频分析装置、服务器及存储介质。通过本发明可以获取视频图像中出现异常事件时的关键信息，以及时处理所述异常事件。

Подробнее

Номер записи: 83

06-08-2014 дата публикации

Signature derivation for images

Номер: CN102124489B

Автор: C·鲍尔, R·拉达克里希南

Принадлежит: Dolby Laboratories Licensing Corp

推导与媒体内容相应的图像指纹涉及选择相同图像的至少两个不同区域、确定这两个区域之间的关系、以及基于这两个图像区域之间的关系推导图像的指纹。

Подробнее

Номер записи: 84

20-07-2015 дата публикации

Method and apparatus for contents play

Номер: KR101537665B1

Автор: 김수웅

Принадлежит: 주식회사 알티캐스트

콘텐츠 재생 방법 및 장치가 개시된다. 콘텐츠를 재생하는 방법은, 부호화된 콘텐츠를 수신하는 단계-상기 콘텐츠는 각 장면별 위치 정보를 포함함; 상기 부호화된 콘텐츠를 복호하는 단계; 상기 복호된 콘텐츠의 각 장면들 중 위치 정보가 달라지는 장면을 추출하는 단계; 상기 추출된 장면에 대한 축소 영상(thumbnail)을 생성하는 단계; 및 상기 축소 영상 및 콘텐츠 중 적어도 하나를 출력하는 단계를 포함한다. A content playback method and apparatus are disclosed. A method of reproducing a content, the method comprising: receiving encoded content, the content including location information for each scene; Decoding the encoded content; Extracting a scene in which positional information is different among scenes of the decoded contents; Generating thumbnails for the extracted scenes; And outputting at least one of the reduced image and the content.

Подробнее

Номер записи: 85

05-01-2021 дата публикации

Flame detection method and device, electronic equipment and storage medium

Номер: CN112001375B

Автор: 吉翔, 周俊琨, 曹亚

Принадлежит: Chengdu Ruiyan Technology Co ltd

本申请提供一种火焰检测方法、装置、电子设备及存储介质，该方法包括：获取视频数据中的待检测图像帧；使用神经网络模型预测出待检测图像帧中火焰所在的位置区域；根据位置区域从视频数据的图像帧中截取对应的区域图像；使用神经网络模型提取截取出的多个区域图像的运动模态特征；使用神经网络模型对运动模态特征进行预测，获得视频数据中是否存在真实火焰的检测结果。在上述的实现过程中，根据多帧视频图像中火焰所在位置的区域图像提取出的运动模态特征来预测，有效地利用了火焰在运动过程中的外观特征与多帧动态运动特征信息，从而提高了对火焰进行检测的准确率。

Подробнее

Номер записи: 86

02-04-2019 дата публикации

Method and system for detecting Internet logo content based on samples and features

Номер: CN109558792A

Автор: 张家亮, 沈宜, 董弋璨, 贾宇, 邹严

Принадлежит: Chengdu 30kaitian Communication Industry Co ltd

本发明公开了一种基于样本和特征检测互联网徽标内容的方法和系统，数据采集单元采集网络上的视图像文件，并将采集的视图像文件进行规范化命名和重复性检验；预处理单元将不同格式的视图像文件转换为统一格式的视图像文件并从视频中抽取出关键帧数据；内容识别分析单元利用卷积神经网络训练样本图集生成特征识别模型，利用特征识别模型对待分析图像数据进行特征相似度分值评估，并进行目标区域计算，获取相应坐标的位置，并提取出视图像特征，从中选择相似度分值最高的一张图片作为标准图，再进行相似度分值评估，对特征图像进行分类；业务逻辑及展示单元对分析后的有害内容进行存储，并形成用户需要的业务逻辑通过网页展示出来。

Подробнее

Номер записи: 87

09-03-2018 дата публикации

Adding method, device, system and the computer-readable medium of audio file

Номер: CN107786549A

Автор: 姚聪

Принадлежит: Beijing Maigewei Technology Co Ltd, Beijing Megvii Technology Co Ltd

本发明提供了一种音频文件的添加方法、装置、系统及计算机可读介质，涉及多媒体信息的技术领域，该方法包括：获取待识别的图像信息；在图像信息中进行手势检测以及识别，以得到手势的手势信息，其中，手势信息包括以下至少之一：手势的位置信息，手势的类型信息，手势的实时持续时间；以及，基于手势信息确定与手势相匹配的音频文件，以在图像信息中添加音频文件，本发明缓解了现有技术中直播视频或者短视频无法基于手势识别来进行音频特效的添加的技术问题。

Подробнее

Номер записи: 88

24-07-2020 дата публикации

Emotion recognition method, emotion recognition device and storage medium

Номер: CN108052889B

Автор: 徐国强, 韦峰

Принадлежит: OneConnect Financial Technology Co Ltd Shanghai

本发明提供一种情绪识别方法、装置及存储介质。该方法包括以下步骤：生成海量中性问题、准绳问题和相关问题，构建测试题库；根据测试题库生成测试问卷；切割测试对象回答测试问卷的视频，得到测试对象回答每个问题的视频片段；提取每个视频片段的表情特征向量，将其视为对应的每个问题的特征向量；计算中性问题的中心点特征向量、准绳问题的中心点特征向量以及每个相关问题的特征向量与中性问题的中心点特征向量的第一距离、每个相关问题的特征向量与准绳问题的中心点特征向量的第二距离；当第一距离大于第二距离时，判定该测试对象隐藏了真实情绪，当第一距离小于第二距离时，判定该测试对象表现出的情绪是真实的。

Подробнее

Номер записи: 89

08-09-2017 дата публикации

Image search method and device

Номер: CN107145487A

Автор: 余登超, 山黎, 常爱平, 张占华, 李平生

Принадлежит: Shenzhen ZTE Netview Technology Co Ltd

本发明公开了一种图像搜索方法和装置，所述方法包括以下步骤：获取用户在输入的图像上选择的至少一个目标区域；提取所述目标区域的图像特征；根据所述目标区域的图像特征搜索图像数据库，从所述图像数据库中获取具有与所述目标区域的图像特征相匹配的区域的图像作为搜索结果。本发明所提供的一种图像搜索方法，允许用户在进行图像搜索时在图像上选取局部区域，并根据局部区域的特征检索图像数据库获取搜索结果，从而只需计算局部区域的特征与图像数据库中的图像的特征进行比对，减少了数据计算量，提高了搜索速度。并且，还实现了图像的局部搜索功能，满足了用户搜索图像局部区域的需求。

Подробнее

Номер записи: 90

27-09-2019 дата публикации

A kind of video object detection method based on attention mechanism

Номер: CN110287826A

Автор: 刘雅琦, 李建强, 白骏

Принадлежит: BEIJING UNIVERSITY OF TECHNOLOGY

本发明涉及一种基于注意力机制的视频目标检测方法，涉及计算机视觉。本发明包括如下步骤：步骤S1，提取当前时间帧的候选特征图；步骤S2，在过去时间段设定融合窗口，计算窗口内各帧的拉普拉斯方差，将方差归一化作为窗口内各帧的权重，将窗口内所有帧的候选特征图进行加权求和得到时序特征，将当前时间帧的候选特征与时序特征相连接，得到待检测特征图；步骤S3，利用卷积层在待检测特征图上提取出额外尺度的特征图；步骤S4，在不同尺度的特征图上利用卷积层进行目标类别及位置预测。本发明的特征融合方法对过去时间段内不同质量的帧特征分配了不同的权重，使得时序信息的融合更加充分，提高了检测模型的性能。

Подробнее

Номер записи: 91

23-03-2006 дата публикации

Hierarchical watermark detector

Номер: US20060062428A1

Автор: Adnan Alattar, Osama Alattar

Принадлежит: Digimarc Corp

The present invention relates generally to digital watermarking. In one implementation, we provide a hierarchical digital watermark detector method. The method includes: i) in a first layer of a hierarchical search, performing watermark detection on blocks of at least a portion of an incoming suspect signal; ii) identifying a first block in the portion that is likely to include a decodable digital watermark; and iii) in a second layer of the hierarchical search, performing additional watermark detection on overlapping blocks in a neighborhood around the first block. Another implementation provides a hierarchical watermark detector including a buffer and a detector. The buffer stores portions of an incoming signal. The detector evaluates watermark detection criteria for blocks stored in the buffer, and hierarchically zooms into a neighborhood of blocks around a block associated with watermark detection criteria that satisfies detection criteria.

Подробнее

Номер записи: 92

18-11-2014 дата публикации

Hierarchical watermark detector

Номер: US8891811B2

Автор: Adnan M. Alattar, Osama M. Alattar

Принадлежит: Digimarc Corp

The present invention relates generally to digital watermarking. In one implementation, we provide a hierarchical digital watermark detector method. The method includes: i) in a first layer of a hierarchical search, performing watermark detection on blocks of at least a portion of an incoming suspect signal; ii) identifying a first block in the portion that is likely to include a decodable digital watermark; and iii) in a second layer of the hierarchical search, performing additional watermark detection on overlapping blocks in a neighborhood around the first block. Another implementation provides a hierarchical watermark detector including a buffer and a detector. The buffer stores portions of an incoming signal. The detector evaluates watermark detection criteria for blocks stored in the buffer, and hierarchically zooms into a neighborhood of blocks around a block associated with watermark detection criteria that satisfies detection criteria.

Подробнее

Номер записи: 93

31-03-2020 дата публикации

Hierarchical watermark detector

Номер: US10607311B2

Автор: Adnan M. Alattar, Osama M. Alattar

Принадлежит: Digimarc Corp

The present invention relates generally to digital watermarking. One aspect of the disclosure includes a method comprising: obtaining data representing imagery; using one or more configured processors, analyzing a plurality of portions of the data to detect a redundantly embedded watermark signal, the analyzing producing detection statistics for each of the plurality of portions, the detection statistics comprising a payload signature, a rotation angle and a scale factor for each portion of the plurality of portions; accumulating payload signatures based on compatible rotation angles and scale factors, said accumulating yielding an accumulated payload signature; and decoding a plural-bit payload from the accumulated payload signature. Of course, many other aspects and disclosure are provided in this patent document.

Подробнее

Номер записи: 94

20-02-2018 дата публикации

Hierarchical watermark detector

Номер: US9898792B2

Автор: Adnan M. Alattar, Osama M. Alattar

Принадлежит: Digimarc Corp

The present invention relates generally to digital watermarking. One aspect of the disclosure includes a method comprising: obtaining data representing imagery; using one or more configured processors, analyzing a plurality of portions of the data to detect a watermark orientation component, said analyzing employing a match filter, in which the match filter yields a correlation value for each of the plurality of portions; determining a first portion from the plurality of portions that comprises a correlation value meeting a predetermined value; and directing a watermark decoder at the first portion to decode a plural-bit watermark payload, in which the watermark decoder produces a watermark signature for the first portion, and in which the watermark decoder searches a plurality of areas at or around the first portion to decode the plural-bit watermark payload. Of course, many other aspects and disclosure are provided in this patent document.

Подробнее

Номер записи: 95

09-08-2016 дата публикации

Hierarchical watermark detector

Номер: US9412144B2

Автор: Adnan M. Alattar, Osama M. Alattar

Принадлежит: Digimarc Corp

The present invention relates generally to digital watermarking. In one implementation, we provide a hierarchical digital watermark detector method. The method includes: i) in a first layer of a hierarchical search, performing watermark detection on blocks of at least a portion of an incoming suspect signal; ii) identifying a first block in the portion that is likely to include a decodable digital watermark; and iii) in a second layer of the hierarchical search, performing additional watermark detection on overlapping blocks in a neighborhood around the first block. Another implementation provides a hierarchical watermark detector including a buffer and a detector. The buffer stores portions of an incoming signal. The detector evaluates watermark detection criteria for blocks stored in the buffer, and hierarchically zooms into a neighborhood of blocks around a block associated with watermark detection criteria that satisfies detection criteria.

Подробнее

Номер записи: 96

22-02-2019 дата публикации

A kind of saliency detection method based on 3D convolutional neural networks

Номер: CN109376611A

Автор: 丁冠群, 方玉明, 杜人钢

Принадлежит: Individual

本发明涉及一种对于视频图像的显著性检测方法，其特征在于：首先利用2D卷积建立一个2D深度卷积神经网络，输入一帧视频帧获得运动目标语义特征，然后利用3D卷积建立一个3D深度卷积神经网络，输入连续三帧视频帧得到时空显著性特征，再将运动目标语义特征和时空显著性信息连接后，输入到3D反卷积网络中，以学习和混合时空显著性特征，最终通过3D反卷积网络得到显著图。这样我们就得到整副图像的显著图，显著值越大，就表明该像素越显著，也就是越吸引人眼关注。实验结果表明我们建立的视频图像显著性检测模型有优良的检测性能。

Подробнее

Номер записи: 97

28-06-2019 дата публикации

AUTOMATIC VIDEO CONTENT Metadata Creation METHOD AND SYSTEM

Номер: KR101994592B1

Автор: 박승보, 윤의녕, 조근식, 홍명덕

Принадлежит: 인하대학교 산학협력단

Disclosed are a method and a system of automatically generating metadata of video content. The system of automatically generating metadata of video content according to the present invention comprises: a video processing unit which performs a preprocessing process of receiving a video to be mutually connected to a script processing unit and recognizing and aligning shot and scene of the video, and analyzes video content; a script processing unit which performs a preprocessing process of receiving a script to be mutually connected to the video processing unit and recognizing and aligning shot and scene, and analyzes script content; an annotation knowledge base which stores annotation metadata generated in the video processing unit; and a narrative knowledge base which stores narrative metadata generated in the script processing unit.

Подробнее

Номер записи: 98

28-09-2016 дата публикации

Video detection method and device

Номер: CN105975939A

Автор: 宋丙玉, 王源, 申晓雷

Принадлежит: Beijing Baidu Netcom Science And Technology Co Ltd

本申请公开了视频检测方法和装置。所述方法的一具体实施方式包括：从待检测视频中截取图像以形成待检测图像序列；生成待检测图像序列中每个待检测图像的哈希签名；将待检测图像序列中各个待检测图像的哈希签名与从样本视频集合中各个样本视频所提取出的样本图像序列中各个样本图像的哈希签名进行匹配；记录样本图像序列中哈希签名匹配成功的样本图像的图像标识以及哈希签名匹配成功的样本图像所属的视频的视频标识，并统计各个图像标识和各个视频标识的出现次数；将视频标识出现次数大于第一阈值且视频标识所关联的图像标识中不同图像标识的出现次数大于第二阈值的样本视频，确定为待检测视频的同源视频。该实施方式实现了同源视频的自动检测。

Подробнее

Номер записи: 99

12-10-2002 дата публикации

Apparatus and method for generating object label images in video sequence

Номер: KR100355382B1

Автор: 김상균, 김지연, 김창용, 문영수, 박두식, 이성덕

Принадлежит: 삼성전자 주식회사

본 발명은 영상 시퀀스에서의 객체 레이블 영상 생성장치 및 그 방법을 개시한다. 각각의 질의 객체를 포함하는 질의 영상들 및 그 각각의 질의 객체를 추출하고자 하는 영상 시퀀스를 입력하고, 영상 시퀀스에서 객체 레이블 영상을 생성하는 장치는, 영상 시퀀스를 유사 장면들의 집합인 일이상의 샷으로 구분하고, 각각의 샷에서 일이상의 대표 프레임을 선택하는 샷 및 대표 프레임 설정부, 질의 객체들에 근거하여 대표 프레임들에서 초기 객체 영역들을 추출하는 초기 객체 영역 추출부, 초기 객체 영역들을 이용하여 샷을 구성하는 모든 프레임들에 대해 객체 영역들을 추적하는 객체 영역 추적부, 및 프레임 단위로 추적된 객체 영역들을 통합하여, 하나의 프레임에서 포함된 일이상의 질의 객체들을 레이블링하는 레이블 영상 생성부를 포함한다. The present invention discloses an apparatus and method for generating an object label image in an image sequence. An apparatus for inputting query images including each query object and an image sequence for extracting each query object, and generating an object label image from the image sequence, converts the image sequence into one or more shots that are sets of similar scenes. A shot and a representative frame setting unit for classifying and selecting one or more representative frames from each shot, an initial object region extraction unit for extracting initial object regions from the representative frames based on query objects, and a shot using the initial object regions An object region tracking unit for tracking the object regions for all the frames constituting a, and a label image generation unit for integrating the object regions tracked in units of frames to label one or more query objects included in one frame.

Подробнее

Номер записи: 100