| Presentation | 2014-03-08 A sign language recognition based on the arm movement and features of hand posture using Kinect Kenta WADA, Naohiro FUKUMURA, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In this study, we examine a recognition of sign language using Kinect which is easily able to measure the arm movement and hand posture. The arm movement is recognized by via-points which are extracted by the minimum-jerk model from hand position obtained from Kinect. The hand posture is recognized by linear classifier using 11 features which are extracted from a hand silhouette image at the start and end point of the sign language's movement. Both recognition results of the arm movement and hand posture are integrated to identify the sign word. We tried recognition experiments using 30 words expressed only by a right hand of 6 subjects. As a result, we got a recognition rate nearly 100% in the single speaker recognition experiment, and nearly 70% in the multiple speakers' recognition experiment. Although it is difficult to improve the recognition rate only by the hand feature, it was shown that the recognition method by integration of the information of hand posture and arm movement has a great improvement in recognition rate. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | sign language recognition / Kinect / arm movement / hand posture / minimum jerk model |
| Paper # | WIT2013-81 |
| Date of Issue |
| Conference Information | |
| Committee | WIT |
|---|---|
| Conference Date | 2014/2/28(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 | Well-being Information Technology(WIT) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | A sign language recognition based on the arm movement and features of hand posture using Kinect |
| Sub Title (in English) | |
| Keyword(1) | sign language recognition |
| Keyword(2) | Kinect |
| Keyword(3) | arm movement |
| Keyword(4) | hand posture |
| Keyword(5) | minimum jerk model |
| 1st Author's Name | Kenta WADA |
| 1st Author's Affiliation | Dept. of Computer Science and Engineering, Toyohashi University of Technology() |
| 2nd Author's Name | Naohiro FUKUMURA |
| 2nd Author's Affiliation | Dept. of Computer Science and Engineering, Toyohashi University of Technology |
| Date | 2014-03-08 |
| Paper # | WIT2013-81 |
| Volume (vol) | vol.113 |
| Number (no) | 481 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |