Presentation | 2001/1/24 Three-dimensional motion estimation for real-time recognition of sign language Tsuyoshi YAMASHIRO, Satoru ODO, Shihoko KAMISATO, Takeshi ARAKAKI, Kiyoshi HOSHINO, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The purpose of this study is to estimate three-dimensional motions of the sign language with a single camera. In the system, a three-dimensional human model with three-dimensional polyhedrons with nine degrees of freedom was prepared, and two-dimensional image sequences with a CCD camera were applied. Moreover, optical flows in three frames before current time were used for the motion estimation, so that the system can be robust to self-occlusions by performer's own face and body. In the experiment, standstill, straight, circular, arc motions and their combinations were adopted for the estimation. As the results, the system realized the gesture estimation and tracking with high accuracy. In the experiment of reproducing three-dimensional motions with computer graphics, gesture reproductions were performed within 0.34ms per a frame without any specific computer hardwares. These results showed that the system is effective for three-dimensional motion estimation for sign language with a single CCD camera. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | sign language / three-dimensional motion estimation / three-dimensional human model / optical flows / self-occlusions / gesture reproduction |
Paper # | EID2000-309,IE2000-164 |
Date of Issue |
Conference Information | |
Committee | EID |
---|---|
Conference Date | 2001/1/24(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Electronic Information Displays (EID) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Three-dimensional motion estimation for real-time recognition of sign language |
Sub Title (in English) | |
Keyword(1) | sign language |
Keyword(2) | three-dimensional motion estimation |
Keyword(3) | three-dimensional human model |
Keyword(4) | optical flows |
Keyword(5) | self-occlusions |
Keyword(6) | gesture reproduction |
1st Author's Name | Tsuyoshi YAMASHIRO |
1st Author's Affiliation | Faculty of Engineering, University of the Ryukyus() |
2nd Author's Name | Satoru ODO |
2nd Author's Affiliation | Faculty of Engineering, University of the Ryukyus |
3rd Author's Name | Shihoko KAMISATO |
3rd Author's Affiliation | Faculty of Engineering, University of the Ryukyus |
4th Author's Name | Takeshi ARAKAKI |
4th Author's Affiliation | Faculty of Engineering, University of the Ryukyus |
5th Author's Name | Kiyoshi HOSHINO |
5th Author's Affiliation | Precursoy Research for Embryonic Science and Technology, Japan Science and Technology Corporation |
Date | 2001/1/24 |
Paper # | EID2000-309,IE2000-164 |
Volume (vol) | vol.100 |
Number (no) | 605 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |