Video multi-scale visualization method and interaction method

Abstract

The invention discloses a video multi-scale visualization method and a video multi-scale interaction method. The video multi-scale interaction method comprises the steps of establishing a user cognition model, which is oriented to a video content structure, of a target video; extracting a foreground object and a background scene in the target video and an image frame of the foreground object; obtaining a moving target and a corresponding trajectory thereof; calculating an appearance density of the moving target according to a time shaft-based moving target appearance amount and a correspondingtime mapping relation; extracting a key frame from processed target video data, and performing annotation on moving target information in the key frame; performing multi-scale division on a processedmoving target identification result and trajectory data of the moving target to generate a multi-scale video information representation structure; and introducing a sketch interactive gesture to an interactive interface of the multi-scale video information representation structure in combination with corresponding semantics of a mouse interactive operation based on an interactive operation mode of a user in an interactive process, and operating the target video on the interactive interface through the sketch interactive gesture.

Description

technical field [0001] The invention belongs to the field of human-computer interaction, and in particular relates to a video multi-scale visualization method and an interaction method. Background technique [0002] Video is an important carrier of information dissemination in today's society. With the rapid development of computer software and hardware technology and the widespread popularization of digital equipment, its data volume has shown an explosive growth trend in daily life. How to quickly obtain video information, and Efficient interaction and effective use of these video information have great application value. With the rapid increase in the number of these video resources, users' access to content-based video data has become more and more widely used, and the demand for presentation, editing, browsing and retrieval of video information has become more and more prominent. However, corresponding to such a huge increase in these video data is that the interaction...

AI Technical Summary

Problems solved by technology

However, in the current research field of surveillance video, the amount of surveillance video data is large, and the content is redundant and cumbersome. When processing surveillance video data, it is difficult for users to quickly locate the surveillance content only through the browsing method based on the time axis. Browse and search (references: Fan C T, Wang Y K, Huang C R. Heterogeneous Information Fusion and Visualization for a Large-Scale Intelligent Video Surveillance System[J]. IEEE Transactions on Systems Man&Cybernetics Systems, 2017,47(4):593- 604.)

Images