Video processing method and device, computer equipment and storage medium

Abstract

The invention discloses a video processing method and device, computer equipment and a storage medium, and the method comprises the steps: for any video, obtaining a content information list of the video, the video comprising M content fragments, and M being a positive integer greater than one; respectively adding the content information list and the video as entities to a knowledge base in an associated manner; and constructing an index based on the content information list entity, and outputting a video entity meeting a retrieval requirement according to the index and a relationship betweenthe content information list and the video. By applying the scheme provided by the invention, the deep retrieval requirement of a user can be met.

Description

[0001] 【Technical field】 [0002] The invention relates to computer application technology, in particular to a video processing method, device, computer equipment and storage medium. [0003] 【Background technique】 [0004] Theatrical gala programs, theatrical performance programs, competitive performance programs (such as I am a singer, poetry conference), etc., these long videos have a common form, that is, the entire program is composed of multiple programs (small programs). The types of the multiple programs may be one or more of songs, dances, cross talk, sketches, and the like. [0005] At present, the materialized long video program information is relatively thin, and usually only includes overall information, such as the name information of the entire program and the duration information of the entire program, etc., and does not include the program information of a single program. Therefore, conventional long video entities Retrieval can only satisfy the user's shallow...

AI Technical Summary

Problems solved by technology

[0005] At present, the materialized long video program information is relatively thin, and usually only includes overall information, such as the name information of the entire program and the duration information of the entire program, etc., and does not include the program information of a single program. Therefore, the conventional long video entity Retrieval can only satisfy the user's shallow search based on the overall information of the long video, but cannot meet the user's deep search needs. For example, if the user retrieves a certain program in a certain long video, the long video cannot be displayed to the user

[0006] At present, there are only program retrieval schemes based on short video forms formed after long video clips. Although some of the in-depth retrieval needs of users are met, the short video information is mixed and the information is not standardized enough to realize the materialization of knowledge graphs. Give full play to the advantages of information association and information accuracy of knowledge graphs. On the contrary, the materialization of long videos has formed a mature product, and the information standardization and authority have been recognized by users, and the video quality is usually good. Therefore, , hoping to replace short videos with long videos and display them to users as search results

Images