| Presentation | 2009-06-11 Hardware Implementation of HOG feature Extraction for Real-time Pedestrian Recognition Ryoji KADOTA, Yukihiro NAKAMURA, Ryusuke MIYAMOTO, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Pedestrian recognition is one of the most challenging problems in the field of computer vision. In this study, feature extraction method using Histograms of Oriented Gradients (HOG) is focused on. HOG is an efficient feature extraction scheme, however, the computational effort of feature extraction is too much. Therefore, this paper proposes a hardware architecture to accelerate calculation of HOG features for real-time pedestrian recognition. The proposed hardware architecture exploits the following two techniques; mathematical functions unsuitable for hardware implementation are replaced with optimal functions, and floating-point numbers are replaced with fixed-point ones. The proposed hardware is synthesized for FPGA. The result shows this hardware performs real-time processing with a small circuit size |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | HOG feature / Histograms of Oriented Gradients / Pedestrian Recognition / Hardware Implementation |
| Paper # | SIS2009-8 |
| Date of Issue |
| Conference Information | |
| Committee | SIS |
|---|---|
| Conference Date | 2009/6/4(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 | Smart Info-Media Systems (SIS) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Hardware Implementation of HOG feature Extraction for Real-time Pedestrian Recognition |
| Sub Title (in English) | |
| Keyword(1) | HOG feature |
| Keyword(2) | Histograms of Oriented Gradients |
| Keyword(3) | Pedestrian Recognition |
| Keyword(4) | Hardware Implementation |
| 1st Author's Name | Ryoji KADOTA |
| 1st Author's Affiliation | Dept. of Communications and Computer Engineering Graduate School of Informatics, Kyoto University() |
| 2nd Author's Name | Yukihiro NAKAMURA |
| 2nd Author's Affiliation | Ritsumeikan University Research Organization of Science and Engineering |
| 3rd Author's Name | Ryusuke MIYAMOTO |
| 3rd Author's Affiliation | Nara Institute of Science and Technology, Graduate School of Information Systems, Communications Laboratory |
| Date | 2009-06-11 |
| Paper # | SIS2009-8 |
| Volume (vol) | vol.109 |
| Number (no) | 78 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |