| Presentation | 2007/1/18 A Study of Omnidirectional Multimodal Estimation for Talker Direction with Equilateral Triangular Microphone Array and Omnidirectional Video Camera Yuki DENDA, Takanobu NISHIURA, Yoichi YAMASHITA, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | This paper proposes an omnidirectional multimodal estimation method for talker direction with equilateral triangular microphone array and omnidirectional video camera. To achieve robust talker direction estimation, audio feature vector is extracted using omnidirectional weighted cross-power spectrum phase (CSP) analysis and CSP coefficient subtraction, and visual feature vector is extracted using normalized distance-based background subtraction and skin color detection medthod in T-S color plane. The talker direction is finally estimated by weighted fusion of audio-visual feature vector based on the reliable criterion of audio feature vector. The results of talker direction estimation experiments in an actual office room revealed that the proposed multimodal talker direction estimation method improved false rejection rate and false acceptance rate than the conventional methods that use audio or visual feature vector, but not both. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Omnidirectional talker direction estimation / multimodal talker direction estimation / feature vector fusion / DOA estimation / Human detection |
| Paper # | TL2006-52,SP2006-140,WIT2006-84 |
| Date of Issue |
| Conference Information | |
| Committee | TL |
|---|---|
| Conference Date | 2007/1/18(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 | Thought and Language (TL) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | A Study of Omnidirectional Multimodal Estimation for Talker Direction with Equilateral Triangular Microphone Array and Omnidirectional Video Camera |
| Sub Title (in English) | |
| Keyword(1) | Omnidirectional talker direction estimation |
| Keyword(2) | multimodal talker direction estimation |
| Keyword(3) | feature vector fusion |
| Keyword(4) | DOA estimation |
| Keyword(5) | Human detection |
| 1st Author's Name | Yuki DENDA |
| 1st Author's Affiliation | Graduate School of Science and Engineering, Ritsumeikan University() |
| 2nd Author's Name | Takanobu NISHIURA |
| 2nd Author's Affiliation | College of Information Science and Engineering, Ritsumeikan University |
| 3rd Author's Name | Yoichi YAMASHITA |
| 3rd Author's Affiliation | College of Information Science and Engineering, Ritsumeikan University |
| Date | 2007/1/18 |
| Paper # | TL2006-52,SP2006-140,WIT2006-84 |
| Volume (vol) | vol.106 |
| Number (no) | 485 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |