Video labeling method and device

Abstract

The invention provides a video labeling method and device in an embodiment. The video labeling method comprises the following steps: obtaining barrage information corresponding to a video, wherein the barrage information contains at least the number of barrage comments at each of different time points when the video is played; determining the time point when the number of barrage comments is greater than or equal to a threshold in the barrage information; determining a video fragment corresponding to the time point; and labeling the video fragment. The labeling method is capable of improving the fineness and accuracy of identifying important content in the video, effectively labeling the video and meeting the requirement of a user on only watching the important video fragment in the video.

Description

technical field [0001] The invention relates to the technical field of video processing, in particular to a video identification method and a video identification device. Background technique [0002] With the rapid development of Internet technology, more and more users search for information or watch videos through the Internet. [0003] Taking watching a video as an example, sometimes the user cannot watch the complete video. For example, when some videos are relatively large and play for a long time, the user may not have time to watch the complete video. In this case, in order to meet the needs of the user, the server may mark the video, and the user only needs to watch the video segment corresponding to the mark. [0004] At present, the video identification method is generally to mark important parts of the entire video for users to locate. However, the existing technology only identifies and filters the obviously unimportant parts of the video, such as the prologue ...

AI Technical Summary

Problems solved by technology

[0004] At present, the video identification method is generally to mark important parts of the entire video for users to locate. However, the existing technology only identifies and filters the obviously unimportant parts of the video, such as the prologue part in TV or movies. and the interspersed advertisements are identified and filtered, and the rest of the video is identified as the key content

This identification method is obviously not fine and accurate enough to identify the important content in the video, and cannot effectively identify the video, and it is difficult to meet the needs of users to only watch important video clips

Images