Video analysis method based on local characteristic descriptor

Abstract

The invention provides a video analysis method based on a local characteristic descriptor. The video analysis method mainly comprises video query, characteristic extraction based on deep learning, compact local characteristic descriptor encoding, video matching and video retrieval, and comprises the steps of firstly extracting a characteristic descriptor of a keyframe in a video, using a color histogram to conduct frame-level distance comparison, combining a manual design characteristic of a compact descriptor used for video analysis and deep learning based on a convolutional neural network, then achieving pair matching through comparison in a coarse-to-precise strategy, finally extracting a candidate keyframe in a database, and conducting sorting through video-grade similarity through further examination on local descriptor matching. In the video analysis method based on the local characteristic descriptor, the redundancy time of the video is eliminated, high-efficiency and low-delay mobile vision search is achieved, the memory size, the bandwidth resource and the cost during running are drastically saved, the compressibility is reduced, and the performance loss is lowered.

Description

technical field [0001] The invention relates to the field of video analysis, in particular to a video analysis method based on local feature descriptors. Background technique [0002] With the development of various emerging technologies, the media information based on pictures and videos increases rapidly, and the related technologies on picture and video processing are also developing rapidly. Among them, video analysis technology has attracted more and more attention. It can be used in mobile AR technology, automobiles, monitoring and media entertainment. In the application of automobiles, video analysis technology has a powerful object detection function, which can effectively warn of collision situations and prompt the current situation. road traffic. In the application of monitoring, once the target violates the predefined rules in the video scene, the system will automatically send out an alarm, and the monitoring workstation will automatically pop up an alarm messag...

AI Technical Summary

Problems solved by technology

[0004] Aiming at problems such as computational complexity and high compression rate, the purpose of the present invention is to provide a video analysis method based on local feature descriptors, first extracting feature descriptors of key frames in the video, and using color histograms to compare frame-level distances , combining manually designed features of compact descriptors for video analysis and deep learning based on convolutional neural networks, and then achieve pairwise matching by comparison in a coarse-to-fine strategy, and finally extract candidate keyframes in the database, through local description Further checks for character matches, sorted by video-level similarity

Images