Video classification method and apparatus, computer device and readable medium

Abstract

The invention provides a video classification method and apparatus, a computer device and a readable medium. The method comprises the steps of obtaining multiple frames of pictures from a target video, identifying each frame of picture according to a pre-trained picture classification model, and predicting a probability of each preset type corresponding to each frame of picture; and according to the probability of each preset type corresponding to each frame of picture, obtaining a target type of the target video. According to the technical scheme, the technical problem of relatively low accuracy of classifying video contents directly based on a video title and a video reply in the prior art can be solved; and by predicting the probability of each preset type corresponding to each frame of picture in the target video based on the pre-trained picture classification model, the target type of the target video is obtained, so that the target video classification accuracy can be effectively improved.

Description

【Technical field】 [0001] The invention relates to the field of computer technology, in particular to a video classification method and device, computer equipment and a readable medium. 【Background technique】 [0002] With the rapid growth of the Internet, it has become an important channel for users to obtain information. But at the same time, the explosion of information has also produced a lot of content that users don't want to see or are not interested in. Especially in recent years, displaying content recommendations based on user interests has gradually become a basic function of Internet products. [0003] For example, in the existing video recommendation scheme in the video field, the videos may be classified according to the title of the video and the user's reply to the video, and so on. Videos are then recommended to users based on the classification of the videos. However, in the prior art, the main sources of editing video titles and video replies are ordinar...

AI Technical Summary

Problems solved by technology

The video replies edited by ordinary users are very random, and the title of the video is more to highlight the mood of the publisher, such as too powerful, very funny, etc., which makes the reply based on the title of the video and the video ineffective. Classify video content, and even be used by some vulgar video content producers

[0004] Therefore, in the prior art, the accuracy of classifying video content based on video titles and video replies is poor

Images