Video abstract method and system used for self-made video
A video summarization and video technology, applied in the field of video summarization, can solve the problems of labor-consuming and time-consuming video summarization, and achieve the effect of saving manpower and time
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0058] This embodiment provides a video summarization method for self-made videos, such as figure 1 shown, including steps:
[0059] S11: Segment the original video using a dense nearest neighbor clustering method to generate video segments;
[0060] S12: Evaluate the importance level of the video segment;
[0061] S13: Sampling video clips within a preset range through a greedy selection strategy.
[0062]Video summarization technology can be applied in real life, such as analyzing surveillance video, video browsing, motion recognition, and generating video logs. It will bring revolutionary technological progress to the fields of transportation and public security. The scientific community and the industry generally believe that the future video system, especially for user-made videos, will use the video summarization system, so this is a technology with broad application prospects, and it will be used in future traffic and public security monitoring equipment. is widely u...
Embodiment 2
[0112] This embodiment provides a video summarization system for self-made videos, such as figure 2 shown, including:
[0113] Segmentation module 21, for using dense nearest neighbor clustering method to segment original video to generate video clips;
[0114] An evaluation module 22, configured to evaluate the importance level of the video segment;
[0115] The generating module 23 is configured to sample video clips within a preset range through a greedy selection strategy.
[0116] This embodiment proposes an unsupervised learning video summarization system based on repeated picture detection and relative representation degree of video frames for user-made videos. This embodiment constructs a system based on a graph model, in which the vertices correspond to video frames, and the edges correspond to the relative entropy of the semantic probability distribution of every two frames, and the video frames or segments are sorted according to the amount of information contain...
PUM

Abstract
Description
Claims
Application Information

- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com