| Presentation | 2011-03-10 Simultaneous Optimization of Context Clustering and GMM for Offline Handwritten Word Recognition Using HMM Tomoyuki HAMAMURA, Bunpei IRIE, Takuya NISHIMOTO, Nobutaka ONO, Shigeki SAGAYAMA, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Context-dependent HMM is commonly used in speech recognition. The model can be realized by two ways : context clustering or tied-mixuture. In speech recognition, the former is reported to be more efficient. However, there is some difficulty in applying context clustering to handwritten word recognition, since the distribution of each character is typically a mixture of some different distributions, such as block-printed, cursive, etc. To deal with this problem, a method for concurrent optimization of context clustering and Gaussian Mixture Model (GMM) is proposed in this paper. Optimization of context clustering by EM algorithm is described first, followed by its expansion to concurrent optimization of context clustering and GMM. The recognition rate of the proposed method is higher than the conventional one which exploits tied-mixture with equivalent computational cost. Experimental results showed 24.2% error reduction on CEDAR database, compared with the conventional tied-mixture based method. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Handwritten word recognition / Context-dependent HMM / Context clustering / GMM / EM algorithm |
| Paper # | PRMU2010-244 |
| Date of Issue |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 2011/3/3(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 | Pattern Recognition and Media Understanding (PRMU) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Simultaneous Optimization of Context Clustering and GMM for Offline Handwritten Word Recognition Using HMM |
| Sub Title (in English) | |
| Keyword(1) | Handwritten word recognition |
| Keyword(2) | Context-dependent HMM |
| Keyword(3) | Context clustering |
| Keyword(4) | GMM |
| Keyword(5) | EM algorithm |
| 1st Author's Name | Tomoyuki HAMAMURA |
| 1st Author's Affiliation | TOSHIBA Corp.:Graduate School of Information Science and Technology, The University of Tokyo() |
| 2nd Author's Name | Bunpei IRIE |
| 2nd Author's Affiliation | TOSHIBA Corp. |
| 3rd Author's Name | Takuya NISHIMOTO |
| 3rd Author's Affiliation | Graduate School of Information Science and Technology, The University of Tokyo |
| 4th Author's Name | Nobutaka ONO |
| 4th Author's Affiliation | Graduate School of Information Science and Technology, The University of Tokyo |
| 5th Author's Name | Shigeki SAGAYAMA |
| 5th Author's Affiliation | Graduate School of Information Science and Technology, The University of Tokyo |
| Date | 2011-03-10 |
| Paper # | PRMU2010-244 |
| Volume (vol) | vol.110 |
| Number (no) | 467 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |