Spotting recognition of human gestures from time-varying images

Author

Nishimura, Takuichi ; Oka, Ryuichi

Author_Institution

Real World Comput. Partnership, Tsukuba Res. Center, Ibaraki, Japan

fYear

1996

fDate

14-16 Oct 1996

Firstpage

318

Lastpage

322

Abstract

We study spotting recognition of human gestures from time-varying images. We propose a feature extraction method and a spotting method. The feature extraction reduces each frame of motion into a frame image of small size, such as 3×3 pixels. We show that the feature with 3×3 is the best one to provide robust characteristics to the changes of cloth and background. The spotting recognition rate was about 80% for 8 gesture categories. A new spotting method called non-monotonic continuous DP is proposed for spotting gestures and their variations such as reverse, partial and stop motions. We show the effectiveness of the new spotting method

Keywords

computer vision; dynamic programming; feature extraction; image recognition; motion estimation; user interfaces; feature extraction method; frame image; human gesture recognition; image motion; nonmonotonic continuous dynamic programming; partial motion; pixels; reverse motions; robust characteristics; spotting method; spotting recognition; stop motion; time-varying images; Animation; Feature extraction; Hidden Markov models; Humans; Image recognition; Image segmentation; Pixel; Robustness; Speech recognition; User interfaces;

fLanguage

English

Publisher

ieee

Conference_Titel

Automatic Face and Gesture Recognition, 1996., Proceedings of the Second International Conference on

Conference_Location

Killington, VT

Print_ISBN

0-8186-7713-9

Type

conf

DOI

10.1109/AFGR.1996.557284

Filename

557284