| Presentation | 2007-06-29 Cascading Approximate Nearest Neighbor Searchers and Its Effects on Object Recognition Kazuto NOGUCHI, Koichi KISE, Masakazu IWAMURA, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | For object recognition based on nearest neighbor search of local descriptors such as SIFT, it is important to make the nearest neighbor search efficient to deal with a huge number of descriptors. In this report we propose a new method of efficient recognition based on the observation that the level of accuracy of nearest neighbor search for correct recognition depends on images to be recognized. The proposed method is characterized by the mechanism that multiple recognizers with approximate nearest neighbor search are cascaded in the order of the level of approximation so as to improve the efficiency by adaptively controlling the level to be applied depending on images. From experimental results for a database of 10,000 images, we have confirmed that the proposed method is capable of achieving a recognition rate of 98% in 1 ms / query, which is about 1/5 of the recognition time without the cascade, and 1/40 of the recognition time with conventional methods of approximate nearest neighbor search such as ANN and LSH. In addition, a recognition error rate of the proposed method has been suppressed to 0.25% by allowing a rejection rate of 12.2%. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Object recognition / SIFT / PCA-SIFT / Approximate nearest neighbor search / Cascade / Reject |
| Paper # | DE2007-18,PRMU2007-44 |
| Date of Issue |
| Conference Information | |
| Committee | DE |
|---|---|
| Conference Date | 2007/6/21(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 | Data Engineering (DE) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Cascading Approximate Nearest Neighbor Searchers and Its Effects on Object Recognition |
| Sub Title (in English) | |
| Keyword(1) | Object recognition |
| Keyword(2) | SIFT |
| Keyword(3) | PCA-SIFT |
| Keyword(4) | Approximate nearest neighbor search |
| Keyword(5) | Cascade |
| Keyword(6) | Reject |
| 1st Author's Name | Kazuto NOGUCHI |
| 1st Author's Affiliation | Graduate School of Engineering, Osaka Prefecture University() |
| 2nd Author's Name | Koichi KISE |
| 2nd Author's Affiliation | Graduate School of Engineering, Osaka Prefecture University |
| 3rd Author's Name | Masakazu IWAMURA |
| 3rd Author's Affiliation | Graduate School of Engineering, Osaka Prefecture University |
| Date | 2007-06-29 |
| Paper # | DE2007-18,PRMU2007-44 |
| Volume (vol) | vol.107 |
| Number (no) | 114 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |