Title :
Quick extracting keyframes from compressed video
Author :
Guo, Xiaojun ; Shi, Fangxia
Author_Institution :
Sch. of Inf. Eng., Tibet Nat. Inst., Xianyang, China
Abstract :
Keyframe is widely used in applications such as non-linear browsing, video content analysis and so on. How to obtain keyframes from video rapidly becomes very important. In this paper, a scheme to extract keyframes from compressed video stream is proposed. It firstly computes the similarity set of adjacent I frames´ DC images, secondly applies clustering algorithm to the similarity set, and finally selects keyframes with the clustering results. The experiment results show that our method is able to extract proper keyframes from test video files fast and easily.
Keywords :
data compression; feature extraction; pattern clustering; video coding; DC images; clustering algorithm; compressed video; keyframe extraction; similarity set; Clustering algorithms; Costs; Data mining; Image analysis; Image coding; Information analysis; Streaming media; Testing; Transform coding; Video compression; DC image; Keyframe extraction; clustering analysis; video compressed domain;
Conference_Titel :
Computer Engineering and Technology (ICCET), 2010 2nd International Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-6347-3
DOI :
10.1109/ICCET.2010.5485659
