Title :
A new video indexing and retrieval method for temporal textures using block-based cooccurrence statistics
Author :
Rahman, Ashfaqur ; Murshed, Manzur ; Dooley, Laurence S.
Author_Institution :
Gippsland Sch. of Comp. & IT, Monash Univ., Churchill, Vic., Australia
Abstract :
A new unified video indexing and retrieval method is presented to classify temporal texture videos using spatial as well as temporal cooccurrence statistics of block-based motion vectors, so keeping the computational complexity for retrieval within a real-time bound. Experimental results clearly demonstrate the superiority of the proposed method over existing temporal cooccurrence matrix-based solutions.
Keywords :
computational complexity; image classification; image motion analysis; image retrieval; image sequences; image texture; matrix algebra; video signal processing; block-based cooccurrence statistics; computational complexity; cooccurrence matrix; motion activity; motion vectors; spatial cooccurrence statistics; temporal cooccurrence statistics; temporal textures; video indexing; video retrieval; video sequences; Australia; Computational complexity; Computer vision; Feature extraction; Focusing; Image segmentation; Indexing; Information retrieval; Statistics; Video sequences;
Conference_Titel :
Networking and Communication Conference, 2004. INCC 2004. International
Print_ISBN :
0-7803-8325-7
DOI :
10.1109/INCC.2004.1366593
