Patent attributes
A video classification method is provided for a computer device. The method includes obtaining a to-be-processed video, where the to-be-processed video has a plurality of video frames, and each video frame corresponds to one time feature; and sampling the to-be-processed video according to a time-feature sampling rule, and obtaining at least one video frame feature sequence. The time-feature sampling rule is a correspondence between time features and video frame feature sequences. The method also includes processing the video frame feature sequence by using a first neural network model, to obtain a feature representation result corresponding to the video frame feature sequence, where the first neural network model is a recurrent neural network model; and processing the feature representation result by using a second neural network model, to obtain a prediction result corresponding to the video frame feature sequence.
