A face video processing method, device, electronic equipment and storage medium

Abstract

The present disclosure relates to a face video processing method, device, electronic device and storage medium, including: acquiring a face video to be processed and a corresponding reference image; using a preset first encoder to extract video frames in the face video to be processed The first feature information is the feature information that needs to be retained in the video frame; the second feature information of the reference image is extracted by using the preset second encoder, and the second feature information is the preset and the face video to be processed. The feature information that matches the processing result; according to the first feature information and the second feature information, the video frames in the face video to be processed are processed to obtain the target face video. In this way, on the one hand, the target face video can be matched with the face video to be processed, and on the other hand, the second feature information is extracted from the reference image, so that the continuity between adjacent video frames in the target face video can also be obtained. Enhancement, thereby improving the stability of the target face video and reducing the possibility of shaking.

Description

technical field [0001] The present disclosure relates to the field of video processing, and in particular to a face video processing method, device, electronic equipment and storage medium. Background technique [0002] With the development of video platforms, video platforms can provide users with more and more functions. In some scenarios, video platforms can provide face video processing functions. Add filters or special effects to the face to meet the needs of users. [0003] In the prior art, the training image and the corresponding real image are obtained first, wherein the real image refers to the real result image after the required transformation of the face in the training video, and then, the training image and the real image are used to compare the preset The model is trained to obtain a face processing model. In this way, when calling the face processing function, the video to be processed can be input to the face video processing model for processing to obtai...

AI Technical Summary

Problems solved by technology

[0004] However, it can be understood that video data is composed of continuous video frames, and there is continuity between adjacent video frames, and the face processing model trained by image data is used to process the video to be processed, which cannot be maintained during the processing process. The continuity between adjacent video frames leads to the lack of continuity and stability of the processed video, and the result of face processing may shake. For example, in the scene where wrinkles are added to the face in the video to be processed, it may cause Although the wrinkle effect is added to each video frame in the processed video, the wrinkle effect between adjacent video frames is quite different, and the processed video is not coherent enough

Images