METHOD AND SYSTEM FOR POSTURAL ANALYSIS AND MEASURING ANATOMICAL DIMENSIONS FROM A RADIOGRAPHIC IMAGE USING MACHINE LEARNING

A method for use of machine learning in computer-assisted anatomical prediction. The method includes identifying with a processor parameters in a plurality of training images to generate a training dataset, the training dataset having data linking the parameters to respective training images, training at least one machine learning algorithm based on the parameters in the training dataset and validating the trained machine learning algorithm, identifying with the processor digitized points on a plurality of anatomical landmarks in a radiographic image of a person's skeleton displayed on a screen by determining anatomical relationships of adjacent bony structures as well as dimensions of at least a portion of a body of the skeleton in the displayed image using the validated machine learning algorithm and a scale factor for the displayed image, and making an anatomical prediction of the person's skeletal alignment based on the determined anatomical dimensions and a known morphological relationship ...

SYSTEM AND METHOD FOR AUTOMATIC LANDMARK LABELING WITH MINIMAL SUPERVISION

A system and method for estimating a set of landmarks for a large image ensemble employs only a small number of manually labeled images from the ensemble and avoids labor-intensive and error-prone object detection, tracking and alignment learning task limitations associated with manual image labeling techniques. A semi-supervised least squares congealing approach is employed to minimize an objective function defined on both labeled and unlabeled images. A shape model is learned on-line to constrain the landmark configuration. A partitioning strategy allows coarse-to-fine landmark estimation.

Image processing method

PROCESS AND DEVICE OF MODELING Of an OBJECT IN an IMAGE

L'invention concerne un procédé de modélisation d'un objet dans une image, utilisant une base d'images, dite base globale, caractérisé en ce que ladite base globale étant partitionnée en bases focalisées, et ladite base globale comportant une base d'initialisation (BI) transverse auxdites bases focalisées, ledit procédé comporte les étapes de:-paramétrage (E1) d'un modèle d'apparence déformable associé à la base d'initialisation (BI) pour l'adapter à ladite image et obtenir des valeurs de paramètres de contrôle (C0),-détermination (E2) d'une ou plusieurs images de la base d'initialisation (BI) pour lesquelles les valeurs des paramètres de contrôle correspondants dudit modèle d'apparence déformable sont les plus proches desdites valeurs de paramètres de contrôle (C0),-détermination (E3) d'une base focalisée (BF) comportant au moins une de ladite ou desdites images de la base d'initialisation (BI),-et paramétrage (E4) d'un modèle d'apparence déformable associé à ladite base focalisée (BF) pour l'adapter à ladite image.

Combining Subcomponent Models for Object Image Modeling

Aspects of the present invention include systems and methods for forming generative models, for utilizing those models, or both. In embodiments, an object model fitting system can be developed comprising a 3D active appearance model (AAM) model. The 3D AAM comprises an appearance model comprising a set of subcomponent appearance models that is constrained by a 3D shape model. In embodiments, the 3D AAM may be generated using a balanced set of training images. The object model fitting system may further comprise one or more manifold constraints, one or more weighting factors, or both. Applications of the present invention include, but are not limited to, modeling and/or fitting face images, although the teachings of the present invention can be applied to modeling/fitting other objects.

RAILWAY TRACK DETECTION DEVICE

Systems and methods for image segmentation

The present disclosure relates to an image processing method. The method may include: obtaining image data; reconstructing an image based on the image data, the image including one or more first edges; obtaining a model, the model including one or more second edges corresponding to the one or more first edges; matching the model and the image; and adjusting the one or more second edges of the model based on the one or more first edges.

FACE TRACKING FOR CONTROLLING IMAGING PARAMETERS

A method of tracking faces in an image stream with a digital image acquisition device includes receiving images from an image stream including faces, calculating corresponding integral images, and applying different subsets of face detection rectangles to the integral images to provide sets of candidate regions. The different subsets include candidate face regions of different sizes and/or locations within the images. The different candidate face regions from different images of the image stream are each tracked.

METHOD FOR GENERATING A THREE-DIMENSIONAL REPRESENTATION OF AN OBJECT

A computer-implemented method of generating a model that models a class of objects. The method comprises, for each of a plurality of objects of the class of objects, receiving a first two- dimensional image of the object and first acquisition data, receiving a second two-dimensional image of the object and second acquisition data and receiving data indicating a plurality of points of interest within the object. The first and second two-dimensional images are combined based upon the first and second acquisition data to generate three-dimensional image data, the three-dimensional image data including data based upon the data indicating a plurality of points of interest and the generated three-dimensional image data for each of the objects of the class of objects is processed to generate the model.

METHOD AND ARRANGEMENT FOR 3D MODEL MORPHING

... 2D 이미지 데이터 입력에 기초하여 표준 3D 모델을 모핑(morphing)하는 방법은, 검출 모델 및 모핑 모델을 이용하여 상기 표준 3D 모델의 초기 모핑을 수행(100)함으로써, 모핑된 표준 3D 모델을 얻는 단계; 상기 2D 이미지 데이터 입력과 상기 모핑된 표준 3D 모델 간의 광학 흐름(optical flow)을 결정하는 단계(200); 상기 광학 흐름을 상기 모핑된 표준 3D 모델에 적용(300)함으로써, 미세 조정 모핑된 3D 표준 모델을 제공하는 단계를 포함한다.

IMAGE SEGMENTATION

Segmentation of foreground from background layers in an image may be provided by a segmentation process which may be based on one or more factors including motion, color, contrast, and the like. Color, motion, and optionally contrast information may be probabilistically fused to infer foreground and/or background layers accurately and efficiently. A likelihood of motion vs. non-motion may be automatically learned from training data and then fused with a contrast-sensitive color model. Segmentation may then be solved efficiently by an optimization algorithm such as a graph cut. © KIPO & WIPO 2008 ...

A method and an apparatus for automatic segmentation of an object

Facial feature processing

Image processing method

A computer-implemented method of generating a model that models a class of objects. The method comprises, for each of a plurality of objects of the class of objects, receiving a first two- dimensional image of the object and first acquisition data, receiving a second two-dimensional image of the object and second acquisition data and receiving data indicating a plurality of points of interest within the object. The first and second two-dimensional images are combined based upon the first and second acquisition data to generate three-dimensional image data, the three-dimensional image data including data based upon the data indicating a plurality of points of interest and the generated three-dimensional image data for each of the objects of the class of objects is processed to generate the model.

Artificial-intelligence-assisted surgery

For Image model construction method and device

LEARNING OBJECT CUTOUT FROM A SINGLE EXAMPLE

Systems and methods are described for learning visual object cutout from a single example. In one implementation, an exemplary system determines the color context near each block in a model image to create an appearance model. The system also learns color sequences that occur across visual edges in the model image to create an edge profile model. The exemplary system then infers segmentation boundaries in unknown images based on the appearance model and edge profile model. In one implementation, the exemplary system minimizes the energy in a graph-cut model where the appearance model is used for data energy and the edge profile is used to modulate edges. The system is not limited to images with nearly identical foregrounds or backgrounds. Some variations in scale, rotation, and viewpoint are allowed.

Image segmentation of foreground from background layers

Segmentation of foreground from background layers in an image may be provided by a segmentation process which may be based on one or more factors including motion, color, contrast, and the like. Color, motion, and optionally contrast information may be probabilistically fused to infer foreground and/or background layers accurately and efficiently. A likelihood of motion vs. non-motion may be automatically learned from training data and then fused with a contrast-sensitive color model. Segmentation may then be solved efficiently by an optimization algorithm such as a graph cut. Motion events in image sequences may be detected without explicit velocity computation.

GENOME-WIDE ASSOCIATION STUDY IDENTIFYING DETERMINANTS OF FACIAL CHARACTERISTICS FOR FACIAL IMAGE GENERATION

The present invention relates to a method for the generation of a facial composite from the genetic profile of a DNA-donor. The method comprises the steps of a) subjecting a biological sample to genotyping thereby generating a profile of genetic markers associated to numerical facial descriptors (NFD) for said sample, b) reverse engineer a NFD from the profile of the associated genetic variants and constructing a facial composite from the reverse engineered numerical facial descriptors (NFDs). The present invention also relates to a method for identifying genetic markers and/or combinations of genetic markers that are predictive of the facial characteristics, (predictive facial markers) of a person, said method comprising the steps of: a) capturing images of a group of individual faces; b) performing image analysis on facial images of said group of individual faces thereby extracting phenotypical descriptors of the faces; c) obtaining data on genetic variation from said group of individual ...

Image processing method

Image processing method and apparatus

An image processing technique includes acquiring a main image of a scene and determining one or more facial regions in the main image. The facial regions are analysed to determine if any of the facial regions includes a defect. A sequence of relatively low resolution images nominally of the same scene is also acquired. One or more sets of low resolution facial regions in the sequence of low resolution images are determined and analysed for defects. Defect free facial regions of a set are combined to provide a high quality defect free facial region. At least a portion of any defective facial regions of the main image are corrected with image information from a corresponding high quality defect free facial region. COPYRIGHT KIPO & WIPO 2010 ...

Method and system for identifying goods of intelligent shopping cart

The present disclosure discloses a method and a system for identifying goods of intelligent shopping cart. The method comprises: reading bar code information of a to-be-purchased goods and obtaining corresponding prestored goods information; continuously detecting and obtaining a total goods weight mn+1in the shopping cart, and comparing the total goods weight mn+1with a total goods weight mnacquired after a previous purchasing action is completed, to obtain a variation mΔ of the total goods weight. According to the method of the present disclosure, when a customer puts a goods in the shopping cart in the shopping course, the correct goods is automatically identified and recorded in a shopping list, then the customer can directly settle the account after completing the shopping, accordingly a lot of time for the customers to wait for the settlement is saved.

IMAGE SEGMENTATION

Segmentation of foreground from background layers in an image may be provided by a segmentation process which may be based on one or more factors including motion, color, contrast, and the like. Color, motion, and optionally contrast information may be probabilistically fused to infer foreground and/or background layers accurately and efficiently. A likelihood of motion vs. non-motion may be automatically learned from training data and then fused with a contrast-sensitive color model. Segmentation may then be solved efficiently by an optimization algorithm such as a graph cut.

Improved age modelling method

본 발명은 얼굴의 화상으로부터 얼굴의 나이 관련 특성들을 모델링하는 방법을 제안하는데, 여기서 나이 관련 특성들은 주름들 또는 나이 반점들 중 하나이고, 그 방법은, - 동일한 성질의 얼굴의 각 나이 관련 특성에 대해, 특성의 모양 및 외모의 파라미터를 포함하는 벡터를 생성하는 단계, - 생성된 벡터로부터 얼굴의 동일한 성질의 나이 관련 특성들을 모델링하는 단일 표현 벡터들을 생성하는 단계를 포함하고, 상기 단일 표현 벡터는 상기 얼굴의 특성들 개수, 및 상기 특성들의 모양 및 외모 특징들의 상기 얼굴에 걸친 결합 확률들을 저장한다. . The present invention proposes a method of modeling age-related characteristics of a face from an image of the face, wherein the age-related characteristics are one of wrinkles or age spots, the method comprising: For each age-related characteristic of the face of the same nature, generating a vector comprising parameters of the shape and appearance of the characteristic, Generating single representation vectors modeling age-related characteristics of the same nature of the face from the generated vector, The single expression vector stores the number of features of the face and the combined probabilities across the face of the shape and appearance features of the features. .

Method and arrangement for 3d model morphing

A method for morphing a standard 3D model based on 2D image data input, comprises the steps of - performing an initial morphing (100) of said standard 3D model using a detection model and a morphing model, thereby obtaining a morphed standard 3D model - determining (200) the optical flow between the 2D image data input and the morphed standard 3D model, - applying (300) the optical flow (300) to said morphed standard 3D model, thereby providing a fine tuned 3D standard model.

Method and apparatus for modelling an object in an image

L'invention concerne un procédé de modélisation d'un objet dans une image, utilisant une base d'images, dite base globale, caractérisé en ce que la base globale étant partitionnée en bases focalisées, et comportant une base d'initialisation (BI) transverse aux bases focalisées, la base d'initialisation et chacune des bases focalisées étant respectivement associées à un modèle d'apparence déformable construit à partir de la base considérée, ce procédé comporte les étapes de : - paramétrage (E1) du modèle d'apparence déformable associé à la base d'initialisation (BI) pour l'adapter à ladite image et obtenir des valeurs de paramètres de contrôle (C 0 ), - détermination (E2) d'une ou plusieurs images de la base d'initialisation (BI) pour lesquelles les valeurs des paramètres de contrôle correspondants du modèle d'apparence déformable associé sont les plus proches desdites valeurs de paramètres de contrôle (C 0 ) obtenues, - détermination (E3) d'une base focalisée (BF) comportant au moins une parmi la ou les images de la base d'initialisation (BI), et - paramétrage (E4) du modèle d'apparence déformable associé à la base focalisée (BF) déterminée, pour l'adapter à ladite image.

IMAGE PROCESSING METHOD AND APPARATUS

IMAGE PROCESSING METHOD

A computer-implemented method of generating a model that models a class of objects. The method comprises, for each of a plurality of objects of the class of objects, receiving a first two- dimensional image of the object and first acquisition data, receiving a second two-dimensional image of the object and second acquisition data and receiving data indicating a plurality of points of interest within the object. The first and second two-dimensional images are combined based upon the first and second acquisition data to generate three-dimensional image data, the three-dimensional image data including data based upon the data indicating a plurality of points of interest and the generated three-dimensional image data for each of the objects of the class of objects is processed to generate the model.

METHOD AND ARRANGEMENT FOR 3D MODEL MORPHING

A method for morphing a standard 3D model based on 2D image data input, comprises the steps of -performing an initial morphing (100) of said standard 3D model using a detection model and a morphing model, thereby obtaining a morphed standard 3D model -determining (200) the optical flow between the 2D image data input and the morphed standard 3D model, -applying (300) the optical flow (300) to said morphed standard 3D model, thereby providing a fine tuned 3D standard model.

System and method for automatic landmark labeling with minimal supervision

A system and method for estimating a set of landmarks for a large image ensemble employs only a small number of manually labeled images from the ensemble and avoids labor-intensive and error-prone object detection, tracking and alignment learning task limitations associated with manual image labeling techniques. A semi-supervised least squares congealing approach is employed to minimize an objective function defined on both labeled and unlabeled images. A shape model is learned on-line to constrain the landmark configuration. A partitioning strategy allows coarse-to-fine landmark estimation.

LEARNING OBJECT CUTOUT FROM A SINGLE EXAMPLE

Systems and methods are described for learning visual object cutout from a single example. In one implementation, an exemplary system determines the color context near each block in a model image to create an appearance model. The system also learns color sequences that occur across visual edges in the model image to create an edge profile model. The exemplary system then infers segmentation boundaries in unknown images based on the appearance model and edge profile model. In one implementation, the exemplary system minimizes the energy in a graph-cut model where the appearance model is used for data energy and the edge profile is used to modulate edges. The system is not limited to images with nearly identical foregrounds or backgrounds. Some variations in scale, rotation, and viewpoint are allowed.

A method and an apparatus for automatic segmentation of an object

The invention relates to a method comprising receiving a plurality of images comprising content that relates to a same object, pre-processing 410 said images to form a feature vector for each region in each image and discovering 420 object-like regions from each image using the feature vectors. An object appearance model is determined for each image according to the object-like regions, an object hypothesis is generated 420 by means of the object appearance model, the same object is segmented 440 in the plurality of images to generate segmented objects and a multiple view segmentation is generated according to the segmented objects. The multiple images may be received from multiple cameras. The preprocessing may comprise using a structure from motion technique in the multiple images to reconstruct sparse 3D points. The invention may be used in a method for capturing 3D content.

Method and apparatus for medical imaging

Method and apparatus for object recognition

A method and an apparatus for object recognition are provided. The method includes: receiving a video including a plurality of frames, and separating the frames into a plurality of frame groups; executing object recognition on a specific frame in each of the frame groups to recognize at least one object in the specific frame; dividing a bonding box of each object into a plurality of sub-blocks, and sampling at least one feature points within at least one of the sub-blocks; and tracking each object in the frames of the frame group according to a variation of the feature points in the frames of the frame group.

Face tracking for controlling imaging parameters

A method of tracking faces in an image stream with a digital image acquisition device includes receiving images from an image stream including faces, calculating corresponding integral images, and applying different subsets of face detection rectangles to the integral images to provide sets of candidate regions. The different subsets include candidate face regions of different sizes and/or locations within the images. The different candidate face regions from different images of the image stream are each tracked.

METHOD AND DEVICE FOR SYNTHESIZING AN IMAGE OF A FACE PARTIALLY OCCLUDED

A method and device for synthesizing a first face in a first image, by determining a first occluded part of the first face that is occluded by an occluding object; determining a first visible part of the first face from the first occluded part; calculating first attributes representative of the first visible part; obtaining first parameters representative of an appearance of the first face by applying a regressor to the first attributes, the regressor modelling a correlation between second attributes representative of second visible parts of a plurality of second faces in second images and second parameters representative of an appearance model of the plurality of second faces; and synthesizing the first face using the first parameters.

3D segmentation for robotic applications

A robotic system includes a robot having an associated workspace; a vision sensor constructed to obtain a 3D image of a robot scene including a workpiece located in the workspace; and a control system communicatively coupled to the vision sensor and to the robot. The control system is configured to execute program instructions to filter the image by segmenting the image into a first image portion containing substantially only a region of interest within the robot scene, and a second image portion containing the balance of the robot scene outside the region of interest; and by storing image data associated with the first image portion. The control system is operative to control movement of the robot to perform work, on the workpiece based on the image data associated with the first image portion.

System and method for automatic landmark labeling with minimal supervision

A system and method for estimating a set of landmarks for a large image ensemble employs only a small number of manually labeled images from the ensemble and avoids labor-intensive and error-prone object detection, tracking and alignment learning task limitations associated with manual image labeling techniques. A semi-supervised least squares congealing approach is employed to minimize an objective function defined on both labeled and unlabeled images. A shape model is learned on-line to constrain the landmark configuration. A partitioning strategy allows coarse-to-fine landmark estimation.

Method and apparatus for medical imaging

The application concerns training a local contour model by providing a training medical scan image to the model, providing a set of local contours for the scan image, providing a set of global universal contours for the scan image, generating using an algorithm an estimated set of universal contours from the set of local contours and the scan image, evaluating a set of differences between the estimated set of universal contours and the set of global universal contours, optimising the algorithm to provide a match between the estimated set of contours and the global universal contours repeating this process with further images until the evaluated set of differences is less than a threshold. The trained model may then be applied to generating an atlas for a universal atlas database as follows: a new medical scan image is provided. A universal auto-contouring operation is performed on the medical scan image, to generate a set of universal contours for the medical scan image. A local auto-contouring ...

Image segmentation

Ultrasound imaging system and method

An ultrasound imaging system is for determining stroke volume and/or cardiac output. The imaging system may include a transducer unit for acquiring ultrasound data of a heart of a subject (or an input for receiving the acquired ultrasound data), and a controller. The controller is adapted to implement a two-step procedure, the first step being an initial assessment step, and the second being an imaging step having two possible modes depending upon the outcome of the assessment. In the initial assessment procedure, it is determined whether regurgitant ventricular flow is present. This is performed using Doppler processing techniques applied to an initial ultrasound data set. If regurgitant flow does not exist, stroke volume is determined using segmentation of 3D ultrasound image data to identify and measure the volume of the left or right ventricle at each of end systole and end-diastole, the difference between them giving a measure of stroke volume. If regurgitant flow does exist, stroke ...

Method and arrangement for image model construction

A method for constructing an image model (M1; M) from at least one image data input (IV1; IV1-IVn), comprises the steps of, in an iterative way , - determining at least one state (PS1; PS1-PSn) of said at least one image data input (IV1; IV1-IVn), and a state (PSMF) of an intermediate learning model (MF; MIF) - determining a target state (TSP) from said at least one state (PS1; PS1-PSn) of said at least one image data input, and from the state (PSMF) of said intermediate learning model (MF; MIF), - performing at least one transformation in accordance with the determined target state (TSP) on said at least one image data input (IV1; IV1-IVn) , thereby generating at least one transformed image (IV1T; IV1 T-IVnT), - aggregating said at least one transformed image (IV1T; IV1T-IVnt) with intermediate learning model (MF; MIF; MIT; MFT) information, thereby generating an updated estimate of said image model (M1; M), - providing said updated estimate of said image model (M1; M) as said image model ...

AUTOMATED METHODS FOR THE OBJECTIVE QUANTIFICATION OF RETINAL CHARACTERISTICS BY RETINAL REGION AND DIAGNOSIS OF RETINAL PATHOLOGY

画像モデルの構築のための方法および装置

Method and apparatus for medical imaging

Genome-wide association study identifying determinants of facial characteristics

The present invention relates to a method for the generation of a facial composite from the genetic profile of a DNA-donor. The method comprises the steps of a) subjecting a biological sample to genotyping thereby generating a profile of genetic markers associated to numerical facial descriptors (NFD) for said sample, b) reverse engineer a NFD from the profile of the associated genetic variants and constructing a facial composite from the reverse engineered numerical facial descriptors (NFDs). The present invention also relates to a method for identifying genetic markers and/or combinations of genetic markers that are predictive of the facial characteristics, (predictive facial markers) of a person, said method comprising the steps of: a) capturing images of a group of individual faces; b) performing image analysis on facial images of said group of individual faces thereby extracting phenotypical descriptors of the faces; c) obtaining data on genetic variation from said group of individual ...

Image processing method

A computer-implemented method of generating a model that models a class of objects. The method comprises, for each of a plurality of objects of the class of objects, receiving a first two- dimensional image of the object and first acquisition data, receiving a second two-dimensional image of the object and second acquisition data and receiving data indicating a plurality of points of interest within the object. The first and second two-dimensional images are combined based upon the first and second acquisition data to generate three-dimensional image data, the three-dimensional image data including data based upon the data indicating a plurality of points of interest and the generated three-dimensional image data for each of the objects of the class of objects is processed to generate the model.

METHOD AND ARRANGEMENT FOR IMAGE MODEL CONSTRUCTION

A method for constructing an image model (M1; M) from at least one image data input (IV1; IV1 -IVn ), comprises the steps of, in an iterative way, - determining at least one state (PS1; PS1 -PSn) of said at least one image data input (IV1; IV1 -IVn), and a state (PSMF) of an intermediate learning model (MF; MIF) - determining a target state (TSP) from said at least one state (PS1; PS1 -PSn) of said at least one image data input, and from the state (PSMF) of said intermediate learning model (MF; MIF), - performing at least one transformation in accordance with the determined target state (TSP) on said at least one image data input (IV1; IV1 -IVn), thereby generating at least one transformed image (IV1T; IV1T-IVnT), - aggregating said at least one transformed image (IV1T; IV1T-IVnt) with intermediate learning model (MF; MIF; MIT; MFT) information, thereby generating an updated estimate of said image model (M1; M), - providing said updated estimate of said image model (M1; M) as said image ...

GENOME-WIDE ASSOCIATION STUDY IDENTIFYING DETERMINANTS OF FACIAL CHARACTERISTICS

System and method for eye alignment in video

A system for image manipulation enables an improved video conferencing experience. The system includes a camera; a display screen adjacent to the camera; a processor coupled to the camera and the display screen; and a memory coupled to the processor. Instructions executable by the processor enable receiving a source image from the camera and generating a synthetic image based upon the source image. The synthetic image corresponds to a view of a virtual camera located at the display screen.

METHOD FOR PROCESSING A REPRESENTAITVE SOURCE IMAGE OF AT LEAST ONE OBJECT, PROCESSING DEVICE, CORRESPONDING DISTANCE MAP AND A COMPUTER SOFTWARE PRODUCT

The invention relates to a method for processing a representative source image (53) of at least one object consisting in building (61) a distance map by associating the smallest distance from the pixel to the outlines of said at least one object to each pixel of the source image (53) according to the outline image obtained at a preliminary stage of searching for at least one outline of said at least one object, in superseding said source image by the distance map at the input of a processing chain, and in processing (63) the distance map (62) in said processing chain.

Method and electronic device for object tracking in a light-field capture

A method and an electronic device for object tracking in a sequence of light-field captures. A data acquisition unit acquires a sequence of light-field captures, wherein each light-field capture comprises a plurality of views. A feature determining unit determines features of an initial visual appearance model for an object of interest in a reference view of a first light-field capture. A feature matching unit matches the features in the reference view and in the further views of the first light-field capture. A feature discarding unit discards features that cannot be well matched in all views of the first light-field capture. An appearance model building unit builds an updated visual appearance model for the object of interest based on the remaining features. Finally, a movement tracking unit tracks the movement of the object of interest in the sequence of light-field captures using the visual appearance model.

Systems and methods for image processing

A method for image segmentation may include acquiring an image including a region of interest (ROI). The ROI has a first margin, the ROI includes a subregion, and the subregion has a second margin. The method may further include acquiring a first model according to the ROI, wherein the first model has a third margin. The method may further determine, based on the first margin and the third margin, a second model by matching the first model with the image, wherein the second model includes a sub-model, and the sub-model has a fourth margin. The method may further include determining, based on the second margin, a third model by adjusting the fourth margin of the sub-model in the second model. The method may further include segmenting the ROI according to the third model and generating a segmented ROI based on a result of the segmentation.

Object detection informed encoding

Embodiments of the present invention provide techniques for coding video data efficiently based on detection of objects within video sequences. A video coder may perform object detection on the frame and when an object is detected, develop statistics of an area of the frame in which the object is located. The video coder may compare pixels adjacent to the object location to the object's statistics and may define an object region to include pixel blocks corresponding to the object's location and pixel blocks corresponding to adjacent pixels having similar statistics as the detected object. The coder may code the video frame according to a block-based compression algorithm wherein pixel blocks of the object region are coded according to coding parameters generating relatively high quality coding and pixel blocks outside the object region are coded according to coding parameters generating relatively lower quality coding.

METHOD OF LOCALIZING LANDMARK POINTS IN IMAGES

A method of localizing landmark points and fitting appearance based models to image data. Image products are computed efficiently which improves the computational cost and improves performance of fitting algorithms for such models.

IMAGE SEGMENTATION

Segmentation of foreground from background layers in an image may be provided by a segmentation process which may be based on one or more factors including motion, color, contrast, and the like. Color, motion, and optionally contrast information may be probabilistically fused to infer foreground and/or background layers accurately and efficiently. A likelihood of motion vs. non-motion may be automatically learned from training data and then fused with a contrast-sensitive color model. Segmentation may then be solved efficiently by an optimization algorithm such as a graph cut.

IMAGE PROCESSING METHOD AND APPARATUS

METHOD AND ARRANGEMENT FOR 3-DIMENSIONAL IMAGE MODEL ADAPTATION

... 객체의 3D 모델(m)을 조정하는 방법으로서, 상기 방법은, - 상기 3D 모델의 적어도 하나의 투영을 수행하여, 관련된 깊이 정보(d1)를 갖는 적어도 하나의 2D 이미지 모델 투영(p1)을 얻는 단계, - 상기 적어도 하나의 2D 이미지 모델 투영(p1)에 대해 적어도 하나의 상태 추출 동작을 수행하여, 적어도 하나의 상태(s1)를 얻는 단계, - 상기 적어도 하나의 상태(s1) 및 타겟 상태(s)에 따라, 상기 적어도 하나의 2D 이미지 모델 투영(p1) 및 상기 관련된 깊이 정보(d1)를 조정함으로써, 적어도 하나의 조정된 2D 이미지 모델(p1') 및 관련된 조정된 깊이(d1')를 얻는 단계, 및 - 상기 관련된 조정된 깊이(d1')에 기초하여, 상기 적어도 하나의 조정된 2D 이미지 모델(p1')을 3D로 역투영함으로써, 조정된 3D 모델(m')을 얻는 단계를 포함하는 방법이 제공된다.

SYSTEM AND METHOD FOR EYE ALIGNMENT IN VIDEO

본 발명의 이미지 조작 시스템은 화상 회의 방식을 개선할 수 있다. 이 시스템은 카메라, 상기 카메라에 인접한 디스플레이 화면, 상기 카메라 및 상기 디스플레이 화면에 연결된 프로세서, 및 상기 프로세서에 연결된 메모리를 포함한다. 프로세서에 의해 실행가능한 명령은 상기 카메라로부터 소스 이미지를 수신하고, 상기 소스 이미지에 기초하여 합성 이미지를 생성하는 것을 가능하게 한다. 상기 합성 이미지는 상기 디스플레이 화면에 배치된 가상 카메라의 뷰에 해당한다. The image manipulation system of the present invention can improve the video conference method. The system includes a camera, a display screen adjacent to the camera, a processor coupled to the camera and the display screen, and a memory coupled to the processor. Instructions executable by the processor to receive a source image from the camera and to generate a composite image based on the source image . The composite image corresponds to a view of a virtual camera disposed on the display screen.

Image processing method and apparatus

Genome-Wide Association Study Identifying Determinants Of Facial Characteristics For Facial Image Generation

The present invention relates to a method for the generation of a facial composite from the genetic profile of a DNA-donor. The method comprises the steps of a) subjecting a biological sample to genotyping thereby generating a profile of genetic markers associated to numerical facial descriptors (NFD) for said sample, b) reverse engineer a NFD from the profile of the associated genetic variants and constructing a facial composite from the reverse engineered numerical facial descriptors (NFDs). The present invention also relates to a method for identifying genetic markers and/or combinations of genetic markers that are predictive of the facial characteristics, (predictive facial markers) of a person, said method comprising the steps of: a) capturing images of a group of individual faces; b) performing image analysis on facial images of said group of individual faces thereby extracting phenotypical descriptors of the faces; c) obtaining data on genetic variation from said group of individual ...

Insassenmodelliervorrichtung, Insassenmodellierverfahren und Insassenmodellierprogramm

Eine Insassenmodelliervorrichtung (30) umfasst einen Beschaffungsabschnitt (34), der ein Bild durch Abbilden eines Bereichs beschafft, wo wahrscheinlich ein Gesicht eines Insassen vorhanden ist, einen Modellanpassungsabschnitt (36), der ein Modell des Gesichts beruhend auf einem durch den Beschaffungsabschnitt beschafften ersten Bild erzeugt, einen Nachführungsabschnitt (38), der das Modell an ein zweites Bild anpasst, das nach dem ersten Bild beschafft wird, einen Bestimmungsabschnitt (40), der eine Richtigkeit einer Gesichtsabschnittsposition, die im zweiten Bild enthalten ist, an das das Modell angepasst ist, unter Verwendung gelernter Informationen bestimmt, die über ein Lernen beruhend auf richtigen Informationen und unrichtigen Informationen hinsichtlich der Gesichtsabschnittsposition erhalten werden, und einen Verarbeitungsabschnitt (32), der gemäß einem Bestimmungsergebnis im Bestimmungsabschnitt bestimmt, ob ein Prozess im Nachführungsabschnitt kontinuierlich auszuführen ist, oder ...

Image processing method

A computer-implemented method of generating a model that models a class of objects is provided. The method comprises, for each of a plurality of objects of the class of objects, receiving a first two- dimensional image of the object and first acquisition data, receiving a second two-dimensional image of the object and second acquisition data and receiving data indicating a plurality of points of interest within the object. The first and second two-dimensional images are combined based upon the first and second acquisition data to generate three-dimensional image data, the three-dimensional image data including data based upon the data indicating a plurality of points of interest and the generated three-dimensional image data for each of the objects of the class of objects is processed to generate the model.

Method and arrangement for 3-dimensional image model adaptation

本發明揭示了一種調整物件的3D模型(m)調整之方法，該方法包含下列步驟：執行該3D模型之至少一投影，以便得到具有相關聯的深度資訊(d1)之至少一2D影像模型投影(p1)；對該至少一2D影像模型投影(p1)執行至少一狀態提取操作，因而得到至少一狀態(s1)；根據該至少一狀態(s1)及一目標狀態(s)而調整該至少一2D影像模型投影(p1)及該相關聯的深度資訊(d1)，因而得到至少一已調整之2D影像模型(p1')及一相關聯的已調整之深度(d1')；以及根據該相關聯的已調整之深度(d1')將該至少一已調整之2D影像模型(p1')反投影到3D，因而得到一已調整之3D模型(m')。

METHOD AND ARRANGEMENT FOR 3-DIMENSIONAL IMAGE MODEL ADAPTATION

IMAGE SEGMENTATION

IMAGE SEGMENTATION

Segmentation of foreground from background layers in an image may be provided by a segmentation process which may be based on one or more factors including motion, color, contrast, and the like. Color, motion, and optionally contrast information may be probabilistically fused to infer foreground and/or background layers accurately and efficiently. A likelihood of motion vs. non-motion may be automatically learned from training data and then fused with a contrast-sensitive color model. Segmentation may then be solved efficiently by an optimization algorithm such as a graph cut.

Method and arrangement for image model construction

A method for constructing an image model (M1; M) from at least one image data input (IV1; IV1-IVn), comprises the steps of, in an iterative way, determining at least one state (PS1; PS1-PSn) of said at least one image data input (IV1; IV1-IVn), and a state (PSMF) of an intermediate learning model (MF; MIF) determining a target state (TSP) from said at least one state (PS1; PS1-PSn) of said at least one image data input, and from the state (PSMF) of said intermediate learning model (MF; MIF), performing at least one transformation in accordance with the determined target state (TSP) on said at least one image data input (IV1; IV1-IVn), thereby generating at least one transformed image (IV1T; IV1T-IVnT), aggregating said at least one transformed image (IV1T; IV1T-IVnt) with intermediate learning model (MF; MIF; MIT; MFT) information, thereby generating an updated estimate of said image model (M1; M), providing said updated estimate of said image model (M1; M) as said image model (M1; M) while ...

Verfahren und Bildverarbeitungsvorrichtung zum Segmentieren von Bilddaten und Computerprogrammprodukt

Die Erfindung betrifft ein Verfahren (15) sowie eine zu dessen Ausführung eingerichtete Bildverarbeitungseinrichtung (1, 3) zum Segmentieren von Bilddaten eines Zielobjekts (2). Bei dem Verfahren (15) wird eine erste Segmentierung mittels eines trainierten Algorithmus (8) erzeugt. Weiter wird ein statistisches Form- und Erscheinungsmodell (10), das auf entsprechende Zielobjekte trainiert ist, bereitgestellt. Weiter wird ein Störungsbereich bestimmt, in dem die Bilddaten durch ein Bildartefakt gestört sind. Es wird dann eine endgültige Segmentierung der Bilddaten durch Anpassen des Form- und Erscheinungsmodells (10) an das jeweilige Zielobjekt (2) außerhalb des Störungsbereiches und Verwenden der ersten Segmentierung in dem Störungsbereich erzeugt.

IMAGE SEGMENTATION

Segmentation of foreground from background layers in an image may be provided by a segmentation process which may be based on one or more factors including motion, color, contrast, and the like. Color, motion, and optionally contrast information may be probabilistically fused to infer foreground and/or background layers accurately and efficiently. A likelihood of motion vs. non-motion may be automatically learned from training data and then fused with a contrast-sensitive color model. Segmentation may then be solved efficiently by an optimization algorithm such as a graph cut. Motion events in image sequences may be detected without explicit velocity computation.

IMAGE SEGMENTATION

PROCESSING COLONOGRAPHY DATA FOR EMPHASIZING LARGE COLORECTAL LESIONS

The invention relates to a method of processing colonography data for emphasizing large colorectal lesions. The method comprises the step of classifying detected lesion candidates as mass candidates before applying a false positive detection reduction step on the detected lesion candidates. Optionally, the classification step comprises a grouping step of grouping lesion candidates that are close to each other.

FACIAL FEATURE PROCESSING

This invention concerns a method, software and system for facial feature processing. It may be applied to face recognition as well as to virtual face synthesis. The method comprises the steps of: Capturing an image including a face in any pose. Applying a cascade face detecting algorithm to the image to find the location of the face in the image. Applying an Active Appearance Model (AAM) search to interpret the face located in the image. Estimating the (horizontal and vertical) orientation of the face. Synthesizing a view of the face from any angle, or using Adaptive Principal Component Analysis (APCA) to a synthesized frontal view of the face for recognition.

Method and arrangement for 3-dimensional image model adaptation

Method for adapting a 3D model (m) of an object, said method comprising the steps of performing at least one projection of said 3D model to obtain at least one 2D image model projection (p1) with associated depth information (d1), performing at least one state extraction operation on said at least one 2D image model projection (p1), thereby obtaining at least one state (s1) adapting said at least one 2D image model projection (p1) and said associated depth information (d1) in accordance with said at least one state (s1) and with a target state (s), thereby obtaining at least one adapted 2D image model (p1′) and an associated adapted depth (d1′) back-projecting said at least one adapted 2D image model (p1′) to 3D, based on said associated adapted depth (d1′) for thereby obtaining an adapted 3D model (m′).

Combining subcomponent models for object image modeling

Aspects of the present invention include systems and methods for forming generative models, for utilizing those models, or both. In embodiments, an object model fitting system can be developed comprising a 3D active appearance model (AAM) model. The 3D AAM comprises an appearance model comprising a set of subcomponent appearance models that is constrained by a 3D shape model. In embodiments, the 3D AAM may be generated using a balanced set of training images. The object model fitting system may further comprise one or more manifold constraints, one or more weighting factors, or both. Applications of the present invention include, but are not limited to, modeling and/or fitting face images, although the teachings of the present invention can be applied to modeling/fitting other objects.

METHOD AND ARRANGEMENT FOR IMAGE MODEL CONSTRUCTION

A method for constructing an image model (M1; M) from at least one image data input (IV1; IV1-IVn), comprises the steps of, in an iterative way, determining at least one state (PS1; PS1-PSn) of said at least one image data input (IV1; IV1-IVn), and a state (PSMF) of an intermediate learning model (MF; MIF) determining a target state (TSP) from said at least one state (PS1; PS1-PSn) of said at least one image data input, and from the state (PSMF) of said intermediate learning model (MF; MIF), performing at least one transformation in accordance with the determined target state (TSP) on said at least one image data input (IV1; IV1-IVn), thereby generating at least one transformed image (IV1T; IV1T-IVnT), aggregating said at least one transformed image (Iv1T; IV1T-IVnt) with intermediate learning model (MF; MIF; MIT; MFT) information, thereby generating an updated estimate of said image model (M1; M), providing said updated estimate of said image model (M1; M) as said image model (M1; M) while ...

Face tracking for controlling imaging parameters

A method of tracking faces in an image stream with a digital image acquisition device includes receiving images from an image stream including faces, calculating corresponding integral images, and applying different subsets of face detection rectangles to the integral images to provide sets of candidate regions. The different subsets include candidate face regions of different sizes and/or locations within the images. The different candidate face regions from different images of the image stream are each tracked.

METHOD AND ARRANGEMENT FOR 3-DIMENSIONAL IMAGE MODEL ADAPTATION

1. Method for adapting a 3D model (m) of an object, said method comprising the steps of - performing at least one projection of said 3D model to obtain at least one 2D image model projection (p1) with associated depth information (d1), - performing at least one state extraction operation on said at least one 2D image model projection (p1), thereby obtaining at least one state (s1) - adapting said at least one 2D image model projection (p1) and said associated depth information (d1) in accordance with said at least one state (s1) and with a target state (s), thereby obtaining at least one adapted 2D image model (p1') and an associated adapted depth (d1') - back-projecting said at least one adapted 2D image model (p1') to 3D, based on said associated adapted depth (d1') for thereby obtaining an adapted 3D model (m').

AUTOMATED METHODS FOR THE OBJECTIVE QUANTIFICATION OF RETINAL CHARACTERISTICS BY RETINAL REGION AND DIAGNOSIS OF RETINAL PATHOLOGY

Automated and objective methods for quantifying a retinal characteristic from a retinal Optometric Coherence Tomography retinal image, and methods for detecting occult ocular pathology, diagnosing ocular pathology, reducing age-bias in OCT image analysis, and monitoring efficacy ocular/retinal disease therapies based on the quantification method are disclosed.

GENOME-WIDE ASSOCIATION STUDY IDENTIFYING DETERMINANTS OF FACIAL CHARACTERISTICS

The present invention relates to a method for the generation of a facial composite from the genetic profile of a DNA-donor. The method comprises the steps of a) subjecting a biological sample to genotyping thereby generating a profile of genetic markers associated to numerical facial descriptors (NFD) for said sample, b) reverse engineer a NFD from the profile of the associated genetic variants and constructing a facial composite from the reverse engineered numerical facial descriptors (NFDs). The present invention also relates to a method for identifying genetic markers and/or combinations of genetic markers that are predictive of the facial characteristics, (predictive facial markers) of a person, said method comprising the steps of: a) capturing images of a group of individual faces; b) performing image analysis on facial images of said group of individual faces thereby extracting phenotypical descriptors of the faces; c) obtaining data on genetic variation from said group of individual ...

Method of Constructing Gray Value or Geometric Models of Anatomic Entity in Medical Image

A gray value model is generated encoding photometric knowledge at landmark positions. This step exploits intensity correlation in neighborhoods sampled around landmark positions. A geometric model is generated encoding geometric knowledge between landmarks. This step exploits spatial correlation between landmarks of segmented anatomic entities.

METHOD AND ARRANGEMENT FOR 3D MODEL MORPHING

Method and apparatus for object recognition

一種物件辨識方法及裝置。所述方法包括：接收包括多個影像幀的影片，並區分這些影像幀為多個影像幀組；對各個影像幀組中的特定影像幀執行物件辨識，以辨識特定影像幀內的至少一個物件；區分各個物件的框選區塊為多個子區塊，並在多個子區塊的至少一子區塊內取樣至少一個特徵點；以及依據特徵點在影像幀組的多個影像幀中的變化追蹤影像幀組的影像幀內的各個物件。

IMAGE PROCESSING METHOD AND APPARATUS

An image processing technique includes acquiring a main image of a scene and determining one or more facial regions in the main image. The facial regions are analysed to determine if any of the facial regions includes a defect. A sequence of relatively low resolution images nominally of the same scene is also acquired. One or more sets of low resolution facial regions in the sequence of low resolution images are determined and analysed for defects. Defect free facial regions of a set are combined to provide a high quality defect free facial region. At least a portion of any defective facial regions of the main image are corrected with image information from a corresponding high quality defect free facial region.

IMAGE PROCESSING METHOD

A computer-implemented method of generating a model that models a class of objects. The method comprises, for each of a plurality of objects of the class of objects, receiving a first two- dimensional image of the object and first acquisition data, receiving a second two-dimensional image of the object and second acquisition data and receiving data indicating a plurality of points of interest within the object. The first and second two-dimensional images are combined based upon the first and second acquisition data to generate three-dimensional image data, the three-dimensional image data including data based upon the data indicating a plurality of points of interest and the generated three-dimensional image data for each of the objects of the class of objects is processed to generate the model.

Face tracking for controlling imaging parameters

A method of tracking faces in an image stream with a digital image acquisition device includes receiving images from an image stream including faces, calculating corresponding integral images, and applying different subsets of face detection rectangles to the integral images to provide sets of candidate regions. The different subsets include candidate face regions of different sizes and/or locations within the images. The different candidate face regions from different images of the image stream are each tracked.

Method and arrangement for 3-Dimensional image model adaptation

Method for adapting a 3D model (m) of an object, said method comprising the steps of - performing at least one projection of said 3D model to obtain at least one 2D image model projection (p1) with associated depth information (d1), - performing at least one state extraction operation on said at least one 2D image model projection (p1), thereby obtaining at least one state (s1) - adapting said at least one 2D image model projection (p1) and said associated depth information (d1) in accordance with said at least one state (s1) and with a target state (s), thereby obtaining at least one adapted 2D image model (p1') and an associated adapted depth (d1') - back-projecting said at least one adapted 2D image model (p1') to 3D , based on said associated adapted depth (d1') for thereby obtaining an adapted 3D model (m').

SYSTEMS AND METHODS FOR IMAGE PROCESSING

A method for image segmentation may include acquiring an image including a region of interest (ROI). The ROI has a first margin, the ROI includes a subregion, and the subregion has a second margin. The method may further include acquiring a first model according to the ROI, wherein the first model has a third margin. The method may further determine, based on the first margin and the third margin, a second model by matching the first model with the image, wherein the second model includes a sub-model, and the sub-model has a fourth margin. The method may further include determining, based on the second margin, a third model by adjusting the fourth margin of the sub-model in the second model. The method may further include segmenting the ROI according to the third model and generating a segmented ROI based on a result of the segmentation.

METHOD, APPARATUS, DEVICE AND STORAGE MEDIUM FOR TRANSFORMING HAIRSTYLE AND COMPUTER PROGRAM PRODUCT

Image segmentation

Segmentation of foreground from background layers in an image may be provided by a segmentation process which may be based on one or more factors including motion, color, contrast, and the like. Color, motion, and optionally contrast information may be probabilistically fused to infer foreground and/or background layers accurately and efficiently. A likelihood of motion vs. non-motion may be automatically learned from training data and then fused with a contrast-sensitive color model. Segmentation may then be solved efficiently by an optimization algorithm such as a graph cut.

METHOD AND ARRANGEMENT FOR IMAGE MODEL CONSTRUCTION

Image segmentation of foreground from background layers

Segmentation of foreground from background layers in an image may be provided by a segmentation process which may be based on one or more factors including motion, color, contrast, and the like. Color, motion, and optionally contrast information may be probabilistically fused to infer foreground and/or background layers accurately and efficiently. A likelihood of motion vs. non-motion may be automatically learned from training data and then fused with a contrast-sensitive color model. Segmentation may then be solved efficiently by an optimization algorithm such as a graph cut.

Methods and apparatus for chatter reduction in video object segmentation using a variable bandwidth search region

Systems, methods, and computer-readable storage media for chatter reduction in video object segmentation using a variable bandwidth search region. A variable bandwidth search region generation method may be applied to a uniform search region to generate a variable bandwidth search region that reduces the search range for segmentation methods such as a graph cut method. The method may identify parts of the contour that are moving slowly, and reduce the search region bandwidth in those places to stabilize the segmentation. This method may determine a bandwidth for each of a plurality of local windows of an image according to an estimate of how much an object in the image has moved from a previous image. The method may blend the bandwidths for the plurality of local windows to generate a blended map. The method may then generate a variable bandwidth search region for an object according to the blended map.

ARTIFICIAL-INTELLIGENCE-ASSISTED SURGERY

Supported by artificial intelligence, an object is classified in an X-ray projection image. A 3D representation as well as a localization of the classified object can be determined by matching a model of the classified object to a visualization of the classified object in the X-ray image. 1. A method for determining a 3D representation and localization of an object in an X-ray projection image with respect to a coordinate system , the method comprising:receiving an X-ray projection image;classifying an object in the X-ray projection image;receiving a model of the classified object;determining a 3D representation of the classified object; andlocalizing the classified object with respect to the coordinate system, by applying the model to match the classified object in the X-ray projection image.2. The method of claim 1 , wherein the X-ray projection image comprises an image characteristic from a group including a pillow effect claim 1 , a curvature claim 1 , noise claim 1 , distortion or a combination thereof claim 1 , the method further comprising accounting for the image characteristic when matching the model with the classified object.3. The method of claim 1 , wherein the classified object comprises an object characteristic selected from a group including X-ray attenuation claim 1 , absorption claim 1 , deflection or a combination thereof claim 1 , the method further comprising accounting for the object characteristic when matching the model with the classified object in the X-ray projection image.4. The method of claim 1 , further comprising detecting at least part of the outline of the classified object in the X-ray projection image.5. The method of claim 1 , further comprising localizing a further object that is not visible or only partially visible in the X-ray image.6. The method of claim 1 , further comprising transferring geometrical aspects from the model to the object.7. The method of claim 6 , further comprising defining a scale based on the geometrical ...

Systems and methods for automatic segmentation in medical imaging with multiple anatomical structure segmentation models

Systems and methods for anatomical structure segmentation in medical images using multiple anatomical structures, instructions and segmentation models.

System and method for eye alignment in video

THREE-DIMENSIONAL PRINTING FROM IMAGES

Methods, systems, and computer readable media for for 3D printing from images, e.g., medical images or images obtained using any appropriate volumetric imaging technology. In some examples, a method includes receiving, from a medical imaging device, a multi-dimensional image of a structure. The method includes, for each two dimensional (2D) slice of the multi-dimensional image, converting, row-by-row for each row of the 2D slice, voxels of the 2D slice into 3D printing instructions for the 2D slice. The method includes 3D printing, by controlling a 3D printing extruder, a physical model based on the structure by 3D printing, slice by slice, each 2D slice using the 3D printing instructions. 1. A method for 3D printing from multi-dimensional images , the method comprising:receiving at least one multi-dimensional image of a structure;for each two dimensional (2D) slice of a plurality of 2D slices of the multi-dimensional image, converting, row-by-row for each row of a plurality of rows of the 2D slice, voxels of the 2D slice into 3D printing instructions for the 2D slice; and3D printing a physical model based on the structure by 3D printing, slice by slice, each 2D slice using the 3D printing instructions.2. The method of claim 1 , wherein receiving the multi-dimensional image of the structure comprises segmenting claim 1 , from the multi-dimensional image claim 1 , a portion of the multi-dimensional image depicting the structure.3. The method of claim 1 , wherein receiving the multi-dimensional image of the structure comprises pre-processing the multi-dimensional image to generate a binary image.4. The method of claim 1 , wherein receiving the multi-dimensional image of the structure comprises resampling the multi-dimensional image to a resolution compatible with a 3D printer.5. The method of claim 1 , wherein receiving the multi-dimensional image of the structure comprises receiving a magnetic resonance (MR) image or a computed tomographic (CT) image of a structure ...

Methods and apparatus for chatter reduction in video object segmentation using optical flow assisted gaussholding

Systems, methods, and computer-readable storage media for chatter reduction in video object segmentation using optical flow assisted gaussholding. An optical flow assisted gaussholding method may be applied to segmentation masks generated for a video sequence. For each frame of at least some frames in a video sequence, for each of one or more other frames prior to and one or more other frames after the current frame, optical flow is computed for the other frame in relation to the current frame and used to warp the contour of the segmentation mask of the other frame to generate warped segmentation mask for the other frames. The weighted average of the warped segmentation masks and the segmentation mask of the current frame is then computed; this weighted average may be blurred spatially, for example using a Gaussian filter. The initial smoothed mask may be thresholded to produce a binary smoothed mask.

Genome-wide association study identifying determinants of facial characteristics for facial image generation

The present invention relates to a method for the generation of a facial composite from the genetic profile of a DNA-donor. The method comprises the steps of a) subjecting a biological sample to genotyping thereby generating a profile of genetic markers associated to numerical facial descriptors (NFD) for said sample, b) reverse engineer a NFD from the profile of the associated genetic variants and constructing a facial composite from the reverse engineered numerical facial descriptors (NFDs). The present invention also relates to a method for identifying genetic markers and/or combinations of genetic markers that are predictive of the facial characteristics, (predictive facial markers) of a person, said method comprising the steps of: a) capturing images of a group of individual faces; b) performing image analysis on facial images of said group of individual faces thereby extracting phenotypical descriptors of the faces; c) obtaining data on genetic variation from said group of individual and d) performing a genome-wide association study (GWAS) to identify said predictive facial markers.

Methods and Apparatus for Chatter Reduction in Video Object Segmentation Using a Variable Bandwidth Search Region

Systems, methods, and computer-readable storage media for chatter reduction in video object segmentation using a variable bandwidth search region. A variable bandwidth search region generation method may be applied to a uniform search region to generate a variable bandwidth search region that reduces the search range for segmentation methods such as a graph cut method. The method may identify parts of the contour that are moving slowly, and reduce the search region bandwidth in those places to stabilize the segmentation. This method may determine a bandwidth for each of a plurality of local windows of an image according to an estimate of how much an object in the image has moved from a previous image. The method may blend the bandwidths for the plurality of local windows to generate a blended map. The method may then generate a variable bandwidth search region for an object according to the blended map.

3d segmentation for robotic applications

A robotic system includes a robot having an associated workspace; a vision sensor constructed to obtain a 3D image of a robot scene including a workpiece located in the workspace; and a control system communicatively coupled to the vision sensor and to the robot. The control system is configured to execute program instructions to filter the image by segmenting the image into a first image portion containing substantially only a region of interest within the robot scene, and a second image portion containing the balance of the robot scene outside the region of interest; and by storing image data associated with the first image portion. The control system is operative to control movement of the robot to perform work, on the workpiece based on the image data associated with the first image portion.

AUTOMATED METHODS FOR THE OBJECTIVE QUANTIFICATION OF RETINAL CHARACTERISTICS BY RETINAL REGION AND DIAGNOSIS OF RETINAL PATHOLOGY

Automated and objective methods for quantifying a retinal characteristic include segmenting an optical coherence tomography retinal image into a plurality of layered retinal regions, and quantifying the retinal characteristic for each region as normalized to a range defined by the characteristic value in the vitreous region and in the retinal pigment epithelium region. Such methods are useful for detecting occult ocular pathology, diagnosing ocular pathology, reducing age-bias in OCT image analysis, and monitoring efficacy ocular/retinal disease therapies. 1. A retinal OCT segmentation algorithm for segmenting a test retinal OCT image into 13 retinal regions , the algorithm comprising:(a) providing a probabilistic shape and intensity model derived from a plurality of OCT images generated from control subjects and segmented into 13 retinal regions;(b) aligning the test OCT image with the probabilistic shape and intensity model according to retinal region shape;(c) comparing pixels of the aligned test OCT image with the corresponding pixels in the control model by intensity, establishing a probable regional fit, and refining regional margins by estimating marginal density distribution with a linear combination a linear combination of discrete Gaussians (LCDG) to generate an initial regional segmentation map of the test OCT image comprising defined retinal regions; and(d) fitting the initial segmentation map with a statistical model that accounts for spatial information to achieve a final segmentation map comprising 13 retinal regions.2. The method according to claim 1 , wherein the alignment of step (b) comprises (i) constructing a wavelet decomposition of the test image claim 1 , (ii) detecting at least three regional boundaries claim 1 , (iii) identifying the foveal pit by reference to the detected boundaries claim 1 , (iv) placing fiduciary points on each detected boundary at locations equidistant from the foveal pit claim 1 , and (v) aligning the test image to the ...

Occupant modeling device, occupant modeling method, and occupant modeling program

An occupant modeling device includes: an acquisition section acquiring an image by imaging a region where there is a probability that a face of an occupant is present; a model fitting section generating a model of the face based on a first image acquired by the acquisition section; a tracking section adapting the model to a second image acquired after the first image; a determination section determining correctness of a facial part position included in the second image to which the model is adapted, by using learned information obtained through learning based on correct information and incorrect information regarding the facial part position; and a processing section determining whether a process in the tracking section is to be continuously executed or a process in the model fitting section is to be executed again according to a determination result in the determination section.

METHOD AND APPARATUS FOR OBJECT RECOGNITION

A method and an apparatus for object recognition are provided. The method includes: receiving a video including a plurality of frames, and separating the frames into a plurality of frame groups; executing object recognition on a specific frame in each of the frame groups to recognize at least one object in the specific frame; dividing a bounded area of each object into a plurality of sub-blocks, and sampling at least one feature point within at least one of the sub-blocks; and tracking each object in the frames in the frame group according to a variation of the feature point in the frames in the frame group. 1. A method for object recognition , applicable to an electronic apparatus that comprises a processor , wherein the method comprises:receiving a video comprising a plurality of frames, and separating the frames into a plurality of frame groups;executing object recognition on a specific frame in each of the frame groups to recognize at least one object in the specific frame;dividing a bounded area of each of the at least one object into a plurality of sub-blocks, and sampling at least one feature point within at least one sub-block of the sub-blocks; andtracking the at least one object in the frames in the frame group according to a variation of the at least one feature point in the frames in the frame group.2. The method according to claim 1 , wherein executing object recognition on the specific frame in each of the frame groups to recognize the at least one object in the specific frame comprises:executing object recognition on a first frame in each of the frame groups to recognize the at least one object in the first frame.3. The method according to claim 1 , wherein sampling the at least one feature point within the at least one sub-block of the sub-blocks comprises:sampling the at least one feature point within a central sub-block located at a center of the sub-blocks.4. The method according to claim 1 , wherein sampling the at least one feature point within ...

SYSTEMS AND METHODS FOR AUTOMATIC SEGMENTATION IN MEDICAL IMAGING WITH MULTIPLE ANATOMICAL STRUCTURE SEGMENTATION MODELS

Systems and methods for anatomical structure segmentation in medical images using multiple anatomical structures, instructions and segmentation models. 1. A network-based system for automatic image segmentation , comprising: access, via the network, a library of different image segmentation models;', 'select and apply all or a subset of the image segmentation models to be used to contour one or more anatomical structures selected by a user via a user interface; and', 'combine results of different segmentation model outcomes., 'a processor configured to2. The system of claim 1 , further comprising:a memory having instructions thereon, wherein when executed, the instructions cause the processor to:access a database including one or more medical images associated with a patient;access a database including the library of different image segmentation models; and determine which ones of the segmentation models in the database support segmentation of at least one of the one or more anatomical structures desired to be contoured;', 'display to the user, on a selection screen, a list of the image segmentation models determined to support segmentation of one or more of the desired anatomical structures, and a list of the anatomical structures that each of the determined segmentation models supports;', 'for each anatomical structure desired to be contoured, allow the user to select one or more segmentation models from the displayed list to be used;', 'apply the selected segmentation models to generate corresponding contours for the desired anatomical structures; and', 'display the generated contours on the one or more images for user review., 'upon a request by the user via the user interface to contour one or more anatomical structures on one or more of the medical images3. The system of claim 2 , wherein the processor is further configured to allow the user to accept or reject any of the generated contours.4. The system of claim 3 , wherein the processor is further configured ...

ULTRASOUND IMAGING SYSTEM AND METHOD

An ultrasound imaging system is for determining stroke volume and/or cardiac output. The imaging system may include a transducer unit for acquiring ultrasound data of a heart of a subject (or an input for receiving the acquired ultrasound data), and a controller. The controller is adapted to implement a two-step procedure, the first step being an initial assessment step, and the second being an imaging step having two possible modes depending upon the outcome of the assessment. In the initial assessment procedure, it is determined whether regurgitant ventricular flow is present. This is performed using Doppler processing techniques applied to an initial ultrasound data set. If regurgitant flow does not exist, stroke volume is determined using segmentation of 3D ultrasound image data to identify and measure the volume of the left or right ventricle at each of end systole and end-diastole, the difference between them giving a measure of stroke volume. If regurgitant flow does exist, stroke volume is determined using Doppler techniques applied to ultrasound data continuously collected throughout a cardiac cycle. 1. An ultrasound imaging system for determining stroke volume of a heart , comprising:an input for receiving ultrasound data of the heart acquired by an ultrasound transducer unit; and perform an initial assessment procedure in which initial ultrasound data is acquired using the transducer unit, and a Doppler processing technique is applied to the initial ultrasound data to determine presence or absence of a regurgitant blood flow from a ventricle to an atrium of a heart captured in the initial ultrasound data;', 'responsive to determining absence of regurgitant flow, implement a first imaging procedure in which: the transducer unit is controlled to acquire 3D ultrasound image data at only an end-systolic time point and end-diastolic time point of a cardiac cycle of the heart, an image segmentation procedure is applied to the 3D ultrasound image data to ...

METHOD AND ELECTRONIC DEVICE FOR OBJECT TRACKING IN A LIGHT-FIELD CAPTURE

A method and an electronic device for object tracking in a sequence of light-field captures. A data acquisition unit acquires a sequence of light-field captures, wherein each light-field capture comprises a plurality of views. A feature determining unit determines features of an initial visual appearance model for an object of interest in a reference view of a first light-field capture. A feature matching unit matches the features in the reference view and in the further views of the first light-field capture. A feature discarding unit discards features that cannot be well matched in all views of the first light-field capture. An appearance model building unit builds an updated visual appearance model for the object of interest based on the remaining features. Finally, a movement tracking unit tracks the movement of the object of interest in the sequence of light-field captures using the visual appearance model. 1. A method for object tracking in a sequence of light-field captures , the method comprising:acquiring a sequence of light-field captures, each light-field capture having a plurality of views;determining features of an initial visual appearance model for an object of interest in a reference view of a first light-field capture;matching the features in the reference view and in the further views of the first light-field capture;discarding features that cannot be well matched in all views of the first light-field capture;building an updated visual appearance model for the object of interest based on the remaining features; andtracking a movement of the object of interest in the sequence of light-field captures using the visual appearance model.2. The method according to claim 1 , wherein the views of each light-field capture are acquired in a timely synchronized manner.3. A non-transitory computer readable storage medium having stored therein instructions enabling object tracking in a sequence of light-field captures claim 1 , which claim 1 , when executed by ...

Method and system for identifying goods of intelligent shopping cart

The present disclosure discloses a method and a system for identifying goods of intelligent shopping cart. The method comprises: reading bar code information of a to-be-purchased goods and obtaining corresponding prestored goods information; continuously detecting and obtaining a total goods weight mn+1 in the shopping cart, and comparing the total goods weight mn+1 with a total goods weight mn acquired after a previous purchasing action is completed, to obtain a variation mΔ of the total goods weight. According to the method of the present disclosure, when a customer puts a goods in the shopping cart in the shopping course, the correct goods is automatically identified and recorded in a shopping list, then the customer can directly settle the account after completing the shopping, accordingly a lot of time for the customers to wait for the settlement is saved.

AGE MODELLING METHOD

Disclosed is a method for modelling age-related traits of a face, from a picture of the face, wherein the age-related traits are either wrinkles or age spots, the method including: for each age-related trait of the face of the same nature, generating a vector including parameters of shape and appearance of the trait; and generating, from the generated vectors, a single representation vector modeling the age-related traits of the same nature in the face. The single representation vector stores information regarding the number of traits in the face and joint probabilities, over the face, of the shape and appearance features of the traits. 2100. A method according to claim 1 , wherein the step () of generating a vector comprising parameters of shape and appearance of the trait comprises:{'b': '110', 'Acquiring coordinates () of a plurality of points of the trait, and'}{'b': '120', 'Processing said coordinates () to infer shape parameters of the trait.'}3. A method according to claim 2 , wherein the age-related traits are wrinkles claim 2 , the method comprising the acquisition of the coordinates of at least five points of each wrinkle claim 2 , wherein the points are regularly spaced over the wrinkle and comprise at least one point on each end of the wrinkle.4. A method according to claim 1 , wherein the shape features of a wrinkle comprise:coordinates of the center of the wrinkle,a length of the wrinkle,and angle of the wrinkle relative to a reference axis, anda curvature of the wrinkle.5130. A method according to claim 1 , wherein the appearance parameters of a wrinkle comprise a thickness σ and depth A of the wrinkle claim 1 , and the processing () of the image to extract these parameters comprises claim 1 , for each wrinkle:high-pass filtering a part of the image comprising the wrinkle,warping each wrinkle into a common shape,fitting a curve on a plurality of transverse profiles of the warped wrinkle to extract, for each profile, a value of a maximum amplitude and ...

Image processing apparatus, image processing method, and computer-readable storing medium

An image processing apparatus includes at least one processor configured to determine whether a mouth in an image of a human face is open or not, on the basis of image information on a central area of the mouth in the face image and image information on a peripheral area of the central area of the mouth in the face image, and correct the image information on the central area of the mouth in the face image in the case where the mouth is open.

Face tracking for controlling imaging parameters

A method of tracking faces in an image stream with a digital image acquisition device includes receiving images from an image stream including faces, calculating corresponding integral images, and applying different subsets of face detection rectangles to the integral images to provide sets of candidate regions. The different subsets include candidate face regions of different sizes and/or locations within the images. The different candidate face regions from different images of the image stream are each tracked. 120.-. (canceled)21. A digital image acquisition device , comprising a lens , an image sensor , a processor , and a processor-readable memory having digital code embedded therein for programming the processor to perform a method of detecting and recognizing faces in digital images acquired by the device , wherein the method comprises:acquiring, from an image stream, an image including one or more face regions;calculating an integral image for at least a portion of the image or a sub-sampled version of the at least a portion of the image, or both;applying a face detection to at least a portion of the integral image to provide one or more candidate face regions;applying a face recognition, which uses a database, to the one or more candidate face regions to recognize a face in the one or more candidate face regions, and determining an identifier for the recognized face; storing the identifier for the recognized face in association with the image or a portion thereof that includes the recognized face; andadjusting one or more image acquisition conditions in an image acquiring device to modify white balance, color balance, focus, or exposure, or combinations thereof, for the recognized face.22. The device of claim 21 , wherein the method further comprises applying a face tracking to detect the recognized face from another image of the image stream.23. The device of claim 21 , wherein the database includes an identifier and one or more associated parameters for ...

Methods and systems for appearance based false positive removal in video analytics

Techniques and systems are provided for maintaining blob trackers for one or more video frames. For example, a blob tracker is identified for a current video frame. The blob tracker is associated with a blob detected for the current video frame. The blob includes pixels of at least a portion of one or more objects in the current video frame. A current characteristic of pixels in a region of the current video frame associated with the blob tracker is determined. A previous characteristic of pixels in a region of a previous video frame associated with the blob tracker is also determined. A difference is determined between the current characteristic and the previous characteristic, and a status of the blob tracker is determined based on the determined difference. The status of the blob tracker indicating whether to maintain the blob tracker for the one or more video frames.

Object detection informed encoding

본 발명의 실시예는 비디오 시퀀스 내의 객체의 검출에 기초하여 비디오 데이터를 효율적으로 코딩하기 위한 기법을 제공한다. 비디오 코더는 프레임에 대해 객체 검출을 수행하고, 객체가 검출되는 경우 객체가 위치되는 프레임의 영역의 통계자료를 생성할 수 있다. 비디오 코더는 객체 위치에 인접한 픽셀들을 객체의 통계자료와 비교할 수 있고, 객체의 위치에 대응하는 픽셀 블록들, 및 검출된 객체와 유사한 통계자료를 갖는 인접한 픽셀들에 대응하는 픽셀 블록들을 포함하는 객체 구역을 정의할 수 있다. 코더는 블록 기반 압축 알고리즘에 따라 비디오 프레임을 코딩할 수 있는데, 여기서 객체 구역의 픽셀 블록들은 상대적으로 높은 품질의 코딩을 발생시키는 코딩 파라미터에 따라 코딩되고, 객체 구역 외부의 픽셀 블록들은 상대적으로 더 낮은 품질의 코딩을 발생시키는 코딩 파라미터에 따라 코딩된다.

Orbit identification device

Method and apparatus for user-customized facial expression recognition

인물의 표정 특징을 효과적으로 표현하면서 조명과 카메라 환경과 같은 외부 요인(artifact) 및 노이즈에 강인한 특징점을 추출함으로써 실시간으로 사람에 독립적인 표정 인식을 가능하게 하는 얼굴 표정 인식 방법 및 장치가 제공된다. 얼굴 표정 인식 방법은 사용자로부터 트레이닝 이미지 시퀀스를 수신하는 단계; 수신된 트레이닝 이미지 시퀀스에 대한 DFEPDM을 학습하며, 학습된 DFEPDM을 이용하여 무표정 이미지를 추출하는 무표정 이미지 추출 단계; 사용자로부터 테스트 이미지 시퀀스를 수신하는 단계; 무표정 이미지 및 테스트 이미지 시퀀스의 AAM) 파라미들 간의 차분치를 이용하여 D-AAM 특징점을 계산하는 D-AAM 특징점 계산 단계; D-AAM 특징점을 학습된 매니폴드 공간으로 투영시켜 차원을 감소시키는 매니폴드 공간 투영 단계; 및 갤러리 시퀀스를 참조하여 매니폴드 공간으로 투영된 D-AAM 특징점들로부터 테스트 이미지 시퀀스의 표정을 인식하는 얼굴 표정 인식 단계를 포함한다. 본 발명에 의하여, 실시간으로 무표정 이미지를 찾고 이를 참조하여 차등-AAM 특징점을 계산할 수 있다.

System and method for automatic segmentation in medical imaging using multiple anatomical segmentation models

用于使用多个解剖结构、指令和分割模型在医学图像中进行解剖结构分割的系统和方法。一种用于自动图像分割的基于网络的系统包括处理器，该处理器被配置为：经由网络访问不同图像分割模型的库；选择和应用图像分割模型的全部或子集以被用于对由用户经由用户界面选择的一个或多个解剖结构进行轮廓化；以及组合不同分割模型效果的结果。

Track recognition device

实施方式的轨道识别装置具备图像取得部、特征量选择部、特征数据导出部及线路检测部。图像取得部取得通过拍摄铁道车辆的行进方向的区域而得到的拍摄图像。特征量选择部针对拍摄图像中的多个像素群的每一个，基于规定的基准，从能够成为表示线路相称度的尺度的多个特征量中选择在线路检测中使用的一个特征量。特征数据导出部从多个像素群中的各个像素群，导出基于由特征量选择部所选择的一个特征量的特征数据。线路检测部基于特征数据导出部的导出结果，从拍摄图像中检测线路。

A method and an apparatus for automatic segmentation of an object

The invention relates to a method, comprising: receiving a plurality of images, wherein the plurality of images comprises content that relates to a same object; preprocessing said more than one images to form a feature vector for each region in an image; discovering object-like regions from each image by means of the feature vectors; determining an object appearance model for each image according to the object-like regions; generating an object hypotheses by means of the object appearance model; segmenting the same object in the plurality of images to generate segmented objects; and generating a multiple view segmentation according to the segmented objects.

Image segmentation

Segmentation of foreground from background layers in an image may be provided by a segmentation process which may be based on one or more factors including motion, color, contrast, and the like. Color, motion, and optionally contrast information may be probabilistically fused to infer foreground and/or background layers accurately and efficiently. A likelihood of motion vs. non-motion may be automatically learned from training data and then fused with a contrast-sensitive color model. Segmentation may then be solved efficiently by an optimization algorithm such as a graph cut.

Image processing method

一种用于产生模拟对象类的模型的计算机实施的方法。该方法包括，针对对象类中的多个对象中的每一者，接收对象的第一二维图像和第一采集数据，接收对象的第二二维图像和第二采集数据，以及接收用于指示对象中多个感兴趣的点的数据。基于第一采集数据和第二采集数据合并第一二维图像和第二二维图像，以产生三维图像数据，该三维图像数据包括基于指示多个感兴趣的点的数据的数据，并且针对对象类中对象中的每一者产生的三维图像数据被处理以产生模型。

The system and method for eye alignment in video

一种用于图像处理的系统能够产生改进的视频会议体验。该系统包括：摄像机；与所述摄像机相邻的显示屏；耦合到所述摄像机和所述显示屏的处理器；以及耦合到所述处理器的存储器。所述处理器可执行的指令能够：从所述摄像机接收源图像；并且基于所述源图像来产生合成图像。所述合成图像与位于所述显示屏处的虚拟摄像机的视图相对应。

Occupant modeling device, occupant modeling method, and occupant modeling program

【課題】トラッキング処理を維持し易い乗員モデリング装置を提供する。【解決手段】乗員モデリング装置は、例えば、車両内において乗員の顔が存在する可能性のある領域を撮像することで得られる画像を取得する取得部と、取得部が取得する第１の画像に基づき顔のモデルを生成するモデルフィッティング部と、取得部が取得する第１の画像より後に取得された第２の画像にモデルを適合させるトラッキング部と、顔の顔部品位置に関する正解情報と不正解情報とに基づき学習させた学習済み情報を用いて、モデルが適合される第２の画像に含まれる顔部品位置の正否を判定する判定部と、判定部の判定結果に応じて、トラッキング部による処理を継続するか、モデルフィッティング部による処理を再実行するかを決定する処理部と、を備える。【選択図】図４

Method, apparatus, device and storage medium for transforming hairstyle and computer program product

The present disclosure discloses a method, apparatus, device, and storage medium for transforming a hairstyle and a computer program product, and relates to the fields of image processing technology, augmented reality technology, and deep learning technology. A specific implementation includes: determining a face bounding box according to information on face key points of acquired face image; constructing grids according to the face bounding box; deforming, by using an acquired target hairstyle function, edge lines of at least a part of the constructed grids, which comprises the hairstyle, to obtain a deformed grid curve; determining a deformed hairstyle in the face image according to the deformed grid curve. The present disclosure employs a target hairstyle function to deform the edge lines of at least part of the grids, which comprises the hairstyle, and does not need to deform the face image as a whole and an output is accurate, and thus the deformed hairstyle in the face image is not distorted.

Improved age modelling method

The invention proposes a method for modelling age-related traits of a face, from a picture of the face, wherein the age-related traits are either wrinkles or age spots, the method comprising: - for each age-related trait of the face of the same nature, generating a vector comprising parameters of shape and appearance of the trait, - generating, from the generated vectors, a single representation vector modeling the age-related traits of the same nature in the face, wherein the single representation vector stores information regarding the number of traits in the face and joint probabilities, over the face, of the shape and appearance features of the traits.

A method, a device, an electronic equipment and a storage medium for changing hairstyle

본 발명은 헤어스타일 변환 방법, 장치, 기기 및 저장 매체에 관하며, 이미지 처리 기술, 증강 현실 기술 및 딥러닝 기술 분야에 관한 것이다. 일 측면에 따른 헤어스타일 변환 방법은, 안면 이미지의 안면 키 포인트 정보에 따라, 안면 바운딩 박스를 결정하는 단계; 안면 바운딩 박스에 따라 그리드를 구축하는 단계; 타깃 헤어스타일 함수를 사용하여 그리드 중 헤어스타일의 적어도 일부 그리드를 포함하는 경계선을 변형하여, 변형된 후의 그리드 곡선을 획득하는 단계; 및 변형된 후의 그리드 곡선에 따라, 안면 이미지 중 변형된 후의 헤어스타일을 결정하는 단계;를 포함한다. 본 발명은 타깃 헤어스타일 함수를 사용하여 그리드 중 헤어스타일의 적어도 일부 그리드를 포함하는 경계선을 변형하여, 안면 이미지 전체를 변환할 필요가 없고 출력이 정확하여, 안면 이미지에서 변형된 후의 헤어스타일이 왜곡되지 않도록 한다. The present invention relates to a hairstyle conversion method, device, device, and storage medium, and relates to the fields of image processing technology, augmented reality technology, and deep learning technology. A hairstyle conversion method according to an aspect includes determining a face bounding box according to face key point information of a face image; building a grid according to the face bounding box; deforming a boundary line including at least a part of the grid of the hairstyle among the grids using the target hairstyle function, and obtaining a grid curve after the deformation; and determining the hair style after transformation in the facial image according to the grid curve after transformation. The present invention uses a target hairstyle function to deform the boundary line including at least a part of the grid of the hairstyle among the grids, so that the entire facial image does not need to be transformed and the output is accurate, so that the hairstyle after being transformed in the facial image is distorted. do not become

Ultrasound imaging system and method

An ultrasound imaging system is for determining stroke volume and/or cardiac output. The imaging system includes a transducer unit for acquiring ultrasound data of a heart of a subject, and a controller. The controller is adapted to implement a two-step procedure, the first step being an initial assessment step, and the second being an imaging step having two possible modes depending upon the outcome of the assessment. In the initial assessment procedure, it is determined whether regurgitant ventricular flow is present. This is performed using Doppler processing techniques applied to an initial ultrasound data set. If regurgitant flow does not exist, stroke volume is determined using segmentation of 3D ultrasound image data to identify and measure the volume of the left or right ventricle at each of end systole and end-diastole, the difference between them giving a measure of stroke volume. If regurgitant flow does exist, stroke volume is determined using Doppler techniques applied to ultrasound data continuously collected throughout a cardiac cycle.

Matching method and device for user 3D modeling

本发明公开了一种用户3D建模的匹配方法和装置，其中，方法包括：向用户面部投射结构光源，并采用第一分辨率拍摄结构光源经过用户面部调制的第一结构光图像；采用图像处理算法对第一结构光图像进行计算，生成用户面部的第一3D模型；应用预设的用户3D模型库对第一3D模型进行匹配，若匹配失败，则采用第二分辨率拍摄结构光源经过用户面部调制的第二结构光图像，其中，第二分辨率大于第一分辨率；采用图像处理算法对第二结构光图像进行计算，生成用户面部的第二3D模型；应用用户3D模型库对第二3D模型进行匹配，显示匹配结果。由此，减小了3D模型匹配的运算量，提高了匹配效率。

Method and system for postural analysis and measuring anatomical dimensions from a radiographic image using machine learning

A method for use of machine learning in computer-assisted anatomical prediction. The method includes identifying with a processor parameters in a plurality of training images to generate a training dataset, the training dataset having data linking the parameters to respective training images, training at least one machine learning algorithm based on the parameters in the training dataset and validating the trained machine learning algorithm, identifying with the processor digitized points on a plurality of anatomical landmarks in a radiographic image of a person's skeleton displayed on a screen by determining anatomical relationships of adjacent bony structures as well as dimensions of at least a portion of a body of the skeleton in the displayed image using the validated machine learning algorithm and a scale factor for the displayed image, and making an anatomical prediction of the person's skeletal alignment based on the determined anatomical dimensions and a known morphological relationship.

Method and apparatus for 3D model morphing

Method and arrangement for image model construction

一种用于从至少一个图像数据输入（IV1；IV1-IVn）来构建图像模型（M1；M）的方法，包括步骤，以迭代的方式，确定所述至少一个图像数据输入（IV1；IV1-IVn）的至少一个状态（PS1；PS1-PSn），以及中间学习模型（MF；MIF）的状态（PSMF）从所述至少一个图像数据输入的所述至少一个状态（PS1；PS1-PSn），以及从所述中间学习模型（MF；MIF）的所述状态（PSMF）来确定目标状态（TSP），根据所述被确定的目标状态（TSP），对所述至少一个图像数据输入（IV1；IV1-IVn）执行至少一个变换，从而生成至少一幅被变换的图像（IV1T；IV1T-IVnT），聚合所述至少一幅被变换的图像（IV1T；IV1T-IVnT）与中间学习模型（MF；MIF；MIT；MFT）信息，从而生成所述图像模型（M1；M）的更新的估计，提供所述图像模型（M1；M）的所述更新的估计作为所述图像模型（M1；M）同时还在反馈循环中，提供所述图像模型（M1；M）的所述更新的估计给模型物体学习模块（500）用于派生所述中间学习模型（MF，MIF）的更新。

Method and apparatus for 3D image model adaptation

Object detection informed encoding

本發明之實施例提供用於基於對視訊序列內之物體的偵測而有效率地編碼視訊資料的技術。一視訊編碼器可對圖框執行物體偵測，且在偵測到一物體時，產生該物體所位於的該圖框之一區域的統計。該視訊編碼器可比較鄰近於該物體位置之像素與該物體之統計，且可界定一物體區以包括對應於該物體位置之像素區塊及對應於具有類似於該經偵測物體之統計的鄰近像素的像素區塊。該編碼器可根據一基於區塊之壓縮演算法來編碼視訊圖框，其中根據產生相對較高品質編碼之編碼參數來編碼該物體區之像素區塊，且根據產生相對較低品質編碼之編碼參數來編碼在該物體區外之像素區塊。

Model-based virtual cleansing for spectral virtual colonoscopy

The present invention relates to assessing the relative contribution of different tissues, or tissue mix. A method is proposed that is based on models of material transitions. The proposed method may be extended to take into account additional information available in increasingly used spectral CT or multi-parametric MRI. Different material concentration maps can be created, which can be used for improved visualization.

Image processing framework for performing object depth estimation

Disclosed are techniques for processing image data. In some aspects, a three-dimensional model can be determined corresponding to an object in an input image. Based on the three-dimensional model, an estimated focal length can be determined that corresponds to the input image. An estimated depth associated with the object in the input image can be calculated based on the estimated focal length and an input image focal length.

Systems and methods for image segmentation

The present disclosure relates to an image processing method. The method may include: obtaining image data; reconstructing an image based on the image data, the image including one or more first edges; obtaining a model, the model including one or more second edges corresponding to the one or more first edges; matching the model and the image; and adjusting the one or more second edges of the model based on the one or more first edges.

Systems and methods for automatic segmentation in medical imaging with multiple anatomical structure segmentation models

Systems and methods for anatomical structure segmentation in medical images using multiple anatomical structures, instructions and segmentation models.

Face pose estimation/three-dimensional face reconstruction method, apparatus, and electronic device

This application discloses methods, apparatus, and electronic devices for face pose estimation and three-dimensional face reconstruction. The face pose estimation method comprises: acquiring a two-dimensional face image for processing, constructing a three-dimensional face model corresponding to the two-dimensional face image, and determining a face pose of the two-dimensional face image based on face feature points of the three-dimensional face model and face feature points of the two-dimensional face image. With this approach, the face pose estimation is performed based on the three-dimensional face model corresponding to the two-dimensional face image, instead of only based on a three-dimensional average face model. As a result, a high accuracy pose estimation can be obtained even for a face with large-angle and exaggerated facial expressions. Thus, robustness of the pose estimation can be effectively improved.

用于分割图像数据的方法和图像处理设备和计算机程序产品

本发明涉及一种方法(15)和一种用于执行该方法的图像处理设备(1，3)，其用于对目标对象(2)的图像数据进行分割。在方法(15)中，借助经训练的算法(8)产生第一分割。此外，提供统计的形状和外观模型(10)，其针对相应的目标对象进行训练。进一步确定图像数据受到图像伪影干扰的干扰区域。然后，通过在干扰区域之外将形状和外观模型(10)与相应的目标对象(2)匹配并且在干扰区域中使用第一分割，来产生图像数据的最终分割。

Three-dimensional printing from images

Methods, systems, and computer readable media for 3D printing from images, e.g., medical images or images obtained using any appropriate volumetric imaging technology. In some examples, a method includes receiving, from a medical imaging device, a multi-dimensional image of a structure. The method includes, for each two dimensional (2D) slice of the multi-dimensional image, converting, row-by-row for each row of the 2D slice, voxels of the 2D slice into 3D printing instructions for the 2D slice. The method includes 3D printing, by controlling a 3D printing extruder, a physical model based on the structure by 3D printing, slice by slice, each 2D slice using the 3D printing instructions.

用于分割图像数据的方法和图像处理设备和计算机程序产品

本发明涉及一种方法(15)和一种用于执行该方法的图像处理设备(1，3)，其用于对目标对象(2)的图像数据进行分割。在方法(15)中，借助经训练的算法(8)产生第一分割。此外，提供统计的形状和外观模型(10)，其针对相应的目标对象进行训练。进一步确定图像数据受到图像伪影干扰的干扰区域。然后，通过在干扰区域之外将形状和外观模型(10)与相应的目标对象(2)匹配并且在干扰区域中使用第一分割，来产生图像数据的最终分割。

用于变换发型的方法、装置、设备以及存储介质

本公开公开了一种用于变换发型的方法、装置、设备以及存储介质，涉及图像处理技术、增强现实技术和深度学习技术领域。具体实现方案为：根据获取的人脸图像的人脸关键点信息，确定人脸包围盒；根据人脸包围盒构建网格；采用获取的目标发型函数对网格中包括发型的至少部分网格的边线进行变形，得到变形后的网格曲线；根据变形后的网格曲线，确定人脸图像中变形后的发型。本公开采用了目标发型函数对网格中包括发型的至少部分网格的边线进行变形，无需对人脸图像整体进行变换且输出准确，使得人脸图像中变形后的发型不失真。

用于机器人应用的3d分割

一种机器人系统，包括：机器人，具有相关联的工作空间；视觉传感器，被构造成获得机器人场景的3D图像，机器人场景包括位于工作空间中的工件；以及控制系统，被通信地耦接到视觉传感器和机器人。控制系统被配置为执行程序指令，以通过将图像分割成第一图像部分和第二图像部分以及通过存储与第一图像部分相关联的图像数据来对图像进行过滤，第一图像部分基本上仅包含机器人场景内的感兴趣区，第二图像部分包含机器人场景的感兴趣区外部的剩余部分。控制系统操作性地基于与第一图像部分相关联的图像数据来控制机器人的移动，以对工件执行作业。

Face pose estimation/three-dimensional face reconstruction method and apparatus, and electronic device

Automated methods for the objective quantification of retinal characteristics by retinal region and diagnosis of retinal pathology

Automated and objective methods for quantifying a retinal characteristic include segmenting an optical coherence tomography retinal image into a plurality of layered retinal regions, and quantifying the retinal characteristic for each region as normalized to a range defined by the characteristic value in the vitreous region and in the retinal pigment epithelium region. Such methods are useful for detecting occult ocular pathology, diagnosing ocular pathology, reducing age-bias in OCT image analysis, and monitoring efficacy ocular/retinal disease therapies.

顔姿勢推定／３次元顔再構築方法、装置、及び電子デバイス

本出願は、顔姿勢推定及び３次元顔再構築のための方法、装置、及び電子デバイスを開示する。顔姿勢推定方法は、処理するための２次元顔画像を取得するステップと、２次元顔画像に対応する３次元顔モデルを構築するステップと、３次元顔モデルの顔特徴点及び２次元顔画像の顔特徴点に基づいて、２次元顔画像の顔姿勢を決定するステップと、を含む。本手法では、顔姿勢推定は、３次元平均顔モデルのみに基づくのではなく、２次元顔画像に対応する３次元顔モデルに基づいて行われる。その結果、角度が大きく表情が誇張された顔に対してさえも、高精度の姿勢推定を得ることができる。したがって、姿勢推定のロバスト性を効果的に向上させることができる。

髪型を変換するための方法、装置、デバイス及び記憶媒体

Image segmentation method and system

图像处理方法、计算机设备及存储介质

本申请公开了一种图像处理方法、计算机设备及存储介质，图像处理方法包括：获取单帧图像内的人物分割结果；基于人物分割结果，查找并标记单帧图像内的连通区域，确定单帧图像内的目标人物区域；通过边界框（bounding box）标记目标人物区域的范围；其中，目标人物区域标记为1，除目标人物区域之外的单帧图像的剩余区域标记为0；保留标记为1的目标人物区域，去除标记为0的剩余区域的背景；替换剩余区域的背景为新的背景。该方法可以消除图像内的无关人员，便于图像背景内容的统一管理，同时对图像背景的处理效率较高。

Face pose estimation/three-dimensional face reconstruction method, apparatus, and electronic device

This application discloses methods, apparatus, and electronic devices for face pose estimation and three-dimensional face reconstruction. The face pose estimation method comprises: acquiring a two-dimensional face image for processing, constructing a three-dimensional face model corresponding to the two-dimensional face image, and determining a face pose of the two-dimensional face image based on face feature points of the three-dimensional face model and face feature points of the two-dimensional face image. With this approach, the face pose estimation is performed based on the three-dimensional face model corresponding to the two-dimensional face image, instead of only based on a three-dimensional average face model. As a result, a high accuracy pose estimation can be obtained even for a face with large-angle and exaggerated facial expressions. Thus, robustness of the pose estimation can be effectively improved.

3d segmentation for robotic applications

A robotic system includes a robot having an associated workspace; a vision sensor constructed to obtain a 3D image of a robot scene including a workpiece located in the workspace; and a control system communicatively coupled to the vision sensor and to the robot. The control system is configured to execute program instructions to filter the image by segmenting the image into a first image portion containing substantially only a region of interest within the robot scene, and a second image portion containing the balance of the robot scene outside the region of interest; and by storing image data associated with the first image portion. The control system is operative to control movement of the robot to perform work, on the workpiece based on the image data associated with the first image portion.

Method and image processing apparatus for the segmentation of image data and computer program product

The disclosure relates to a method and to an image processing facility configured to carry out the method for the segmentation of image data of a target object. In the method, a first segmentation is generated by a trained algorithm. Furthermore, a statistical shape and appearance model is provided, which is trained on corresponding target objects. An interference region is further determined, in which the image data is impaired by an image artifact. A final segmentation of the image data is then generated by adjusting the shape and appearance model to the respective target object outside the interference region and using the first segmentation in the interference region.

Cerebrovascular segmentation from mra images

There is provided a method of processing a cerebrovascular medical image, the method comprising receiving magnetic resonance angiography (MRA) image associated with a cerebrovascular tissue comprising blood vessels and brain tissues other than blood vessels; segmenting MRA image using a prior appearance model for generating first prior appearance features representing a first-order prior appearance model and second appearance features representing a second-order prior appearance model of the cerebrovascular tissue, wherein current appearance model comprises a 3D Markov-Gibbs Random Field (MGRF) having a 2D rotational and translational symmetry such that MGRF model is 2D rotation and translation invariant; segmenting MRA image using current appearance model for generating current appearance features distinguishing blood vessels from other brain tissues; adjusting MRA image using first and second prior appearance features and current appearance futures; and generating an enhanced MRA image based on said adjustment. There is also provided a system for doing the same.

Segmentation of medical images

Methods for segmenting medical images from different modalities include integrating a plurality of types of quantitative image descriptors with a deep 3D convolutional neural network. The descriptors include: (i) a Gibbs energy for a prelearned 7th-order Markov-Gibbs random field (MGRF) model of visual appearance, (ii) an adaptive shape prior model, and (iii) a first-order appearance model of the original volume to be segmented. The neural network fuses the computed descriptors to obtain the final voxel-wise probabilities of the goal regions.

Generating combined diagnostic imaging and pathology images

A computer-implemented image generation method (100) for generating combined diagnostic imaging and pathology images is described. The method (100) comprises: acquiring with a first imaging modality pathological image data (110) of a subject; acquiring with a second imaging modality diagnostic image data (120) of the subject; mapping (140) the pathological image data to the diagnostic image data; and applying an image generation algorithm (150) in order to generate a combined diagnostic and pathology image based on the mapping of the data.

从单个示例学习对象剪切

描述了用于从单个示例学习视觉对象剪切的系统和方法。在一个实现中，示例性系统确定模式图像中每个块附近的色彩上下文以创建外观模型。该系统还学习跨模型图像中的视觉边缘出现的色彩序列以创建边缘轮廓模型。该示例性系统接着基于外观模型和边缘轮廓模型来推断未知图像中的分割边界。在一个实现中，该示例性系统将图形剪切模型中的能量最小化，其中外观模型用于数据能量而边缘轮廓用于调整边缘。该系统不限于具有几乎相同的前景或背景的图像。比例、旋转和视点上的某些变化是允许的。

用于从单个示例学习对象剪切的方法和系统

本发明描述了用于从单个示例学习视觉对象剪切的系统和方法。在一个实现中，示例性系统确定模式图像中每个块附近的色彩上下文以创建外观模型。该系统还学习跨模型图像中的视觉边缘出现的色彩序列以创建边缘轮廓模型。该示例性系统接着基于外观模型和边缘轮廓模型来推断未知图像中的分割边界。在一个实现中，该示例性系统将图形剪切模型中的能量最小化，其中外观模型用于数据能量而边缘轮廓用于调整边缘。该系统不限于具有几乎相同的前景或背景的图像。比例、旋转和视点上的某些变化是允许的。

Artificial-intelligence-assisted surgery

Supported by artificial intelligence, an object is classified in an X-ray projection image, and by matching a model of the classified object to a visualization of the classified object in the X-ray image a 3D representation as well as a localization of the classified object can be determined.

图像处理方法、计算机设备及存储介质

本申请公开了一种图像处理方法、计算机设备及存储介质，图像处理方法包括：获取单帧图像内的人物分割结果；基于人物分割结果，查找并标记单帧图像内的连通区域，确定单帧图像内的目标人物区域；通过边界框（bounding box）标记目标人物区域的范围；其中，目标人物区域标记为1，除目标人物区域之外的单帧图像的剩余区域标记为0；保留标记为1的目标人物区域，去除标记为0的剩余区域的背景；替换剩余区域的背景为新的背景。该方法可以消除图像内的无关人员，便于图像背景内容的统一管理，同时对图像背景的处理效率较高。

用于利用多个解剖结构分割模型在医学成像中进行自动分割的系统和方法

用于使用多个解剖结构、指令和分割模型在医学图像中进行解剖结构分割的系统和方法。一种用于自动图像分割的基于网络的系统包括处理器，该处理器被配置为：经由网络访问不同图像分割模型的库；选择和应用图像分割模型的全部或子集以被用于对由用户经由用户界面选择的一个或多个解剖结构进行轮廓化；以及组合不同分割模型效果的结果。

用于分割图像数据的方法和图像处理设备和计算机程序产品

本发明涉及一种方法(15)和一种用于执行该方法的图像处理设备(1，3)，其用于对目标对象(2)的图像数据进行分割。在方法(15)中，借助经训练的算法(8)产生第一分割。此外，提供统计的形状和外观模型(10)，其针对相应的目标对象进行训练。进一步确定图像数据受到图像伪影干扰的干扰区域。然后，通过在干扰区域之外将形状和外观模型(10)与相应的目标对象(2)匹配并且在干扰区域中使用第一分割，来产生图像数据的最终分割。

用于变换发型的方法、装置、设备以及存储介质

本公开公开了一种用于变换发型的方法、装置、设备以及存储介质，涉及图像处理技术、增强现实技术和深度学习技术领域。具体实现方案为：根据获取的人脸图像的人脸关键点信息，确定人脸包围盒；根据人脸包围盒构建网格；采用获取的目标发型函数对网格中包括发型的至少部分网格的边线进行变形，得到变形后的网格曲线；根据变形后的网格曲线，确定人脸图像中变形后的发型。本公开采用了目标发型函数对网格中包括发型的至少部分网格的边线进行变形，无需对人脸图像整体进行变换且输出准确，使得人脸图像中变形后的发型不失真。

轨道识别装置

实施方式的轨道识别装置具备图像取得部、特征量选择部、特征数据导出部及线路检测部。图像取得部取得通过拍摄铁道车辆的行进方向的区域而得到的拍摄图像。特征量选择部针对拍摄图像中的多个像素群的每一个，基于规定的基准，从能够成为表示线路相称度的尺度的多个特征量中选择在线路检测中使用的一个特征量。特征数据导出部从多个像素群中的各个像素群，导出基于由特征量选择部所选择的一个特征量的特征数据。线路检测部基于特征数据导出部的导出结果，从拍摄图像中检测线路。

超声成像系统和方法

一种超声成像系统用于确定搏出量和/或心输出量。所述成像系统可以包括：换能器单元，其用于采集对象的心脏的超声数据(或用于接收所采集的超声数据的输入)；以及控制器。所述控制器适于执行两步流程，第一步是初始评估步骤，第二步是具有取决于评估结果两种可能模式的成像步骤。在初始评估流程中，确定是否存在返流性心室流量。这是使用应用于初始超声数据的多普勒处理技术来执行的。如果不存在返流，则使用3D超声图像数据分割来确定搏出量，以识别和测量收缩末期和舒张末期中的每个中左心室或右心室的体积，它们之间的差值可以给出搏出量的量度。如果确实存在返流，则使用多普勒技术确定搏出量，多普勒技术可应用于整个心动周期连续收集的超声数据。

人脸姿态估计/三维人脸重构方法、装置及电子设备

本申请公开了人脸姿态估计方法和装置，三维人脸重构方法和装置，以及电子设备。其中，人脸姿态估计方法包括：获取待处理的二维人脸图像，构建所述二维人脸图像对应的三维人脸模型，根据所述三维人脸模型的人脸特征点和所述二维人脸图像的人脸特征点，确定所述二维人脸图像的人脸姿态。采用这种处理方式，使得基于与二维人脸图像对应的三维人脸模型进行姿态求解，而不再只基于三维平均脸模型进行姿态求解，使得在对人脸大角度、夸张表情进行姿态估计时，仍可获得较高精准度的姿态；因此，可以有效提升姿态估计的鲁棒性。

人脸姿态估计/三维人脸重构方法、装置及电子设备

本申请公开了人脸姿态估计方法和装置，三维人脸重构方法和装置，以及电子设备。其中，人脸姿态估计方法包括：获取待处理的二维人脸图像，构建所述二维人脸图像对应的三维人脸模型，根据所述三维人脸模型的人脸特征点和所述二维人脸图像的人脸特征点，确定所述二维人脸图像的人脸姿态。采用这种处理方式，使得基于与二维人脸图像对应的三维人脸模型进行姿态求解，而不再只基于三维平均脸模型进行姿态求解，使得在对人脸大角度、夸张表情进行姿态估计时，仍可获得较高精准度的姿态；因此，可以有效提升姿态估计的鲁棒性。