Compressed domain action classification using HMM

Author

Babu, Venkatesh R. ; Anantharaman, B. ; Ramakrishnan, K.R. ; Srinivasan, S.H.

Author_Institution

Dept. of Electr. Eng., Indian Inst. of Sci., Bangalore, India

fYear

2001

fDate

2001

Firstpage

44

Lastpage

49

Abstract

The paper proposes three techniques for person independent action classification in compressed MPEG video. The features used are based on motion vectors, obtained by partial decoding of the MPEG video. The features proposed are projected 1D, 2D polar and 2D Cartesian. The feature vectors are fed to a hidden Markov model (HMM) for classification of actions. In total, seven actions were trained with distinct HMM for classification. Recognition results of more than 90% have been achieved. This work is significant in the context of the emerging MPEG-7 standard for video indexing and retrieval

Keywords

data compression; decoding; feature extraction; hidden Markov models; image classification; video coding; 2D Cartesian; 2D polar; HMM; MPEG-7 standard; compressed MPEG video; compressed domain action classification; feature vectors; hidden Markov model; motion vectors; partial decoding; person independent action classification; video indexing; video retrieval; Decoding; Electronic mail; Feature extraction; Hidden Markov models; Image coding; Indexing; Linearity; MPEG 7 Standard; Transform coding; Video compression;

fLanguage

English

Publisher

ieee

Conference_Titel

Content-Based Access of Image and Video Libraries, 2001. (CBAIVL 2001). IEEE Workshop on

Conference_Location

Kauai, HI

Print_ISBN

0-7695-1354-9

Type

conf

DOI

10.1109/IVL.2001.990855

Filename

990855