| Presentation | 2011-12-16 A hierarchical extention of the HOG model implemented in the convolution-net for human detection Yasuto ARAKAKI, Hayaru SHOUNO, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In the field of the image recognition, HOG model, which is a feature extractor based on the local gradient information of the image, is proposed for a detection of person in images. The HOG model makes a good performance for the pedestrian detection ever though its simple detection mechanism. However, we consider the HOG model has several problems as following: the size of extracted feature dimensions may become large by settings of image dividing parameters. In addition the location and the scale invariances are not satisfied the HOG model. In order to overcome these problems, we proposed new model that introduce a concept of the convolution net which was a model of the visual processing system in the brain of the mammals. In order to evaluate of the performance of our proposing model, we use the INRIAPerson Data Set which is pedestrian detection database, and we discussed about recognition performance. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | HOG model / Conbolution-net / Hierarchical HOG model |
| Paper # | PRMU2011-132 |
| Date of Issue |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 2011/12/8(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) | A hierarchical extention of the HOG model implemented in the convolution-net for human detection |
| Sub Title (in English) | |
| Keyword(1) | HOG model |
| Keyword(2) | Conbolution-net |
| Keyword(3) | Hierarchical HOG model |
| 1st Author's Name | Yasuto ARAKAKI |
| 1st Author's Affiliation | The Univversity of Electro-Communications() |
| 2nd Author's Name | Hayaru SHOUNO |
| 2nd Author's Affiliation | The Univversity of Electro-Communications |
| Date | 2011-12-16 |
| Paper # | PRMU2011-132 |
| Volume (vol) | vol.111 |
| Number (no) | 353 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |