Video processing method and apparatus, computer device, and storage medium

A video processing apparatus, comprising:

• at least one memory;
• and
• one or more processors, coupled to the at least one memory;
• wherein one or more application programs are stored in the at least one memory and configured to be executed by the one or more processors, and the one or more application programs being configured to implement:
• obtaining a first video frame and a second video frame in a target video, the first video frame being a previous frame of the second video frame;
• inputting the first video frame and the second video frame to a target neural network, and obtaining a target intermediate video frame output by the target neural network, the target neural network being obtained by training a preset neural network based on an optical flow distillation constraint and a feature consistency constraint, the feature consistency constraint being established between training intermediate frame features and true sample intermediate frame features, the training intermediate frame features being generated by a decoder of the preset neural network, the true sample intermediate frame features corresponding to a plurality of levels of the decoder of the preset neural network, the true sample intermediate frame features being obtained by inputting a true sample intermediate frame to a parameter sharing encoder that shares parameters with an encoder of the preset neural network;
• and
• interpolating the target intermediate video frame between the first video frame and the second video frame.

A non-transitory computer-readable storage medium, storing program codes, the program codes being invoked by at least one processor to implement:

• obtaining a first video frame and a second video frame in a target video, the first video frame being a previous frame of the second video frame;
• inputting the first video frame and the second video frame to a target neural network, and obtaining a target intermediate video frame output by the target neural network, the target neural network being obtained by training a preset neural network based on an optical flow distillation constraint and a feature consistency constraint, the feature consistency constraint being established between training intermediate frame features and true sample intermediate frame features, the training intermediate frame features being generated by a decoder of the preset neural network, the true sample intermediate frame features corresponding to a plurality of levels of the decoder of the preset neural network, the true sample intermediate frame features being obtained by inputting a true sample intermediate frame to a parameter sharing encoder that shares parameters with an encoder of the preset neural network;
• and
• interpolating the target intermediate video frame between the first video frame and the second video frame.