Video generation method and system for high resolution face swapping

A video generation system for high resolution face swapping, wherein the video generation system comprises:

• an image obtaining device, configured to obtain a target face image in a to-be-processed video and a source face image corresponding to the target face image;
• a face feature encoder, configured to extract a feature of each of the source face image and the target face image, to obtain source feature codes corresponding to the source face image and target feature codes corresponding to the target face image;
• a face feature exchanger, configured to generate swapped face feature codes according to the source feature codes and the target feature codes;
• a face generator, configured to generate an initial swapped face image according to the swapped face feature codes;
• and
• a face fuser, configured to fuse the initial swapped face image with the target face image, to obtain a final swapped face image.

A video generation method for high resolution face swapping, comprising:

• a processor;
• and
• a memory, configured to store computer-executable instructions, wherein the executable instructions, when executed, cause the processor to perform the following operations:
• obtaining a target face image in a to-be-processed video and a source face image corresponding to the target face image;
• extracting a feature of each of the source face image and the target face image through a face feature encoder, to obtain source feature codes corresponding to the source face image and target feature codes corresponding to the target face image;
• generating swapped face feature codes through a face feature exchanger according to the source feature codes and the target feature codes;
• generating an initial swapped face image through a face generator according to the swapped face feature codes;
• and
• fusing the initial swapped face image with the target face image through a face fuser, wherein the face fuser is based on face semantic segmentation, to obtain a final swapped face image.

A non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores on or more programs, and the one or more programs, when executed by an electronic device comprising a plurality of application programs, cause the electronic device to perform the following operations:

• obtaining a target face image in a to-be-processed video and a source face image corresponding to the target face image;
• extracting a feature of each of the source face image and the target face image through a face feature encoder, to obtain source feature codes corresponding to the source face image and target feature codes corresponding to the target face image;
• generating swapped face feature codes through a face feature exchanger according to the source feature codes and the target feature codes;
• generating an initial swapped face image through a face generator according to the swapped face feature codes;
• and
• fusing the initial swapped face image with the target face image through a face fuser, wherein the face fuser is based on face semantic segmentation, to obtain a final swapped face image.