基于双向自适应时空切片的关键帧提取算法  

Key Frame Extraction Algorithm Based on Bidirectional Adaptive Spatio-Temporal Slice

作　　者：韩震博 HAN Zhen-bo(College of Computer Science,Sichuan University,Chengdu 610065)

机构地区：[1]四川大学计算机学院,成都610065

出　　处：《现代计算机》2021年第5期57-61,共5页Modern Computer

摘　　要：关键帧提取是视频分析、视频检索等领域的重要研究内容,针对现有关键帧提取算法中常出现的查全率低、冗余帧多等问题,提出一种基于双向自适应时空切片的关键帧提取算法。实验结果表明,该算法性能优良,提取出的关键帧能够较为准确地表达视频的主要内容。Key frame extraction is an important research content in video analysis,video retrieval and other fields.Aiming at the problems of low re⁃call rate and many redundant frames in existing key frame extraction algorithms,a key frame extraction algorithm based on bidirectional adaptive Spatio-Temporal Slice is proposed.Experimental results show that the algorithm has good performance,and the extracted key frames can accurately express the main content of the video.

关 键 词：关键帧提取 自适应 时空切片 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]