Rapid method of key frame extraction for MPEG compressed video
Wang Chang-cheng · Jisuanji gongcheng yu sheji · 2005
Key frame extraction is a fundamental processes for content-base video retrieval and video analysis.Key frame can dramatically reduce the data of the video indexing,and provide a framework for video abstraction and retrieval.In this paper,at first briefly discuss the existing approaches,and then present a motion feature based key frame extracting method using fuzzy reasoning theory for MPEG compressed video.By performing minimal decoding on compressed bit stream,computational complexity is reduced,and the processing speed can be highly improved.Experiments show that this proposed approach is efficient and can make a good representation for video contents.