Video abstract generation method and device based on shortest path

Abstract

The invention discloses a video abstract generation method and a video abstract generation device based on the shortest path. The method comprises the steps of acquiring video frame information of a video to be processed; sampling the video frame information, and acquiring sampling information of the video frame information; determining a segmentation position of the video to be processed according to the sampling information, and segmenting the video to be processed at the segmentation position to acquire a segmented video; clustering video frames in the segmented video, and acquiring a candidate key frame; and building a digraph, computing the shortest path through combining the digraph with the shortest path algorithm, eliminating redundancy of the candidate key frame of the shortest path to acquire a key frame, and generating a video abstract according to the key frame. The method and device provided by the invention have the advantages of simple graph model, small calculation amount and high video abstract acquisition efficiency.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a shortest-path-based video summary generation method and device. Background technique [0002] Video summarization, also known as video enrichment, is a simple summary of video content. Whether it is daily life or professional needs, it is often necessary to quickly browse and retrieve massive video data, and generating a summary of the video to display the key content of the video can greatly improve the efficiency of video browsing and retrieval. [0003] Static video summarization is a form of video summarization, also known as video summary, which represents the content of the video with a series of static semantic units extracted from the original video stream. Static semantic units are such as key frames, titles, slides, etc. Slices and the like can generally represent the static feature information of the video shot content. Typical clustering-based met...

AI Technical Summary

Problems solved by technology

[0004] These traditional methods based on clustering methods and graphical models have been verified to have certain effectiveness, but in fact they all have their own defects.

For example, in the clustering method, it is often necessary to comprehensively consider issues such as the initial cluster center setting, the determination of the number of key frames, and the stability of threshold selection. Although there is no need to set the cluster center and the number of key frames in advance, it has a large computational complexity. , the disadvantage that it is difficult to select the merge point or the split point; in addition, the key frame extraction is affected by subjective factors to a certain extent

However, in the actual situation of the method based on the graphical model, the number of video frames of a video is often quite large. At the same time, the relationship between all the frame pictures needs to be considered, the graphical model will become quite complicated, and the amount of calculation will also increase.

Images