Настройки

Укажите год
-

Небесная энциклопедия

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

Подробнее
-

Мониторинг СМИ

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

Подробнее

Форма поиска

Поддерживает ввод нескольких поисковых фраз (по одной на строку). При поиске обеспечивает поддержку морфологии русского и английского языка
Ведите корректный номера.
Ведите корректный номера.
Ведите корректный номера.
Ведите корректный номера.
Укажите год
Укажите год
Применить Всего найдено 2. Отображено 2.
21-07-2023 дата публикации

Construction method of movable object detection model, detection model and detection method

Номер: CN116468984A
Автор: CHEN WENHUI, YANG YANQING, ZHAO HUIHUANG, ZHENG GUANGYONG, LIANG XIAOMAN, LIN MUGANG, WAN XIAOQING
Принадлежит:

The invention belongs to the technical field of target detection, and particularly relates to a construction method of a movable object detection model, the detection model and a detection method. According to the method, distribution calculation is performed on marked samples in a training data image to generate a distribution matrix, the distribution matrix is encoded according to a preset distribution density, and then a target identification model is utilized to predict the marked samples to obtain the confidence coefficient, the prediction number and the prediction center coordinate of a to-be-identified movable object; the method comprises the following steps: firstly, marking center coordinates of a to-be-identified movable object, then mapping the predicted center coordinates of the to-be-identified movable object to a distribution matrix to obtain a region density code corresponding to the predicted center coordinates, then generating a prediction result mark according to the overlapping ...

Подробнее
Номер записи: 1
02-05-2023 дата публикации

Method for changing face after makeup based on deep convolutional neural network and face key point detection

Номер: CN116052240A
Автор: ZHAO HUIHUANG, ZHANG FACHAO, LIANG XIAOMAN, LIN MUGANG, SUN YAQI, WANG YAONAN
Принадлежит:

The invention discloses an after-makeup face changing method based on a deep convolutional neural network and face key point detection, and the method comprises the steps: firstly, selecting a target portrait image and an after-makeup portrait image which need to be converted, then carrying out the face key point detection of the two images, employing a Dlib frame to detect the face key points, and carrying out the correction of the obtained key points, then storing the key point coordinates of the human face in the two pictures; using a BiSeNet network model to obtain a face contour; using a StyleBlit algorithm to generate a picture after face changing; and finally, obtaining a final makeup face changing image by combining the eye and mouth selection areas generated by face segmentation with the selection area of the complete face. According to the invention, face key point detection is carried out on the portrait image after makeup and the target portrait image, face changing after makeup ...

Подробнее
Номер записи: 2