Video feature extraction and video content understanding method, apparatus, storage medium and server
Abstract:
Provided are a video processing method and apparatus, a video retrieval method and apparatus, a medium, and a server. The video processing method includes: performing encoding and decoding on an original video by using the encoder and the decoder, to obtain a video feature of the original video and a hidden state of the original video at a decoding stage; reconstructing a video feature of a target video by using the reconstructor according to the hidden state of the original video at the decoding stage; obtaining a difference between the video feature of the target video and the video feature of the original video; and adjusting a processing parameter of at least one of the decoder and the reconstructor to reduce the difference between the video feature of the target video and the video feature of the original video.
Information query
Patent Agency Ranking
0/0