| Presentation | 2021-08-26 Pedestrian Position Estimation by Integrating AlphaPose Based Human Pose Estimation and PredRNN++ Based Spatiotemporal Series Prediction for Autonomous Driving Systems Zhaowei Zhang, Eiji Fukuzawa, Jun Ohya, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In recent years, with the development of automatic driving technology, more experimental automatic driving vehicles have been developed, where it is inevitable to solve the problem of security. In automatic driving systems, it is an important research topic to predict how the accidence will occur in the future. The key of the problem is to predict the position of pedestrians in front of the vehicle. Pedestrian position prediction under the risk prediction branch is the issue to be solved in this research. By comparing several related studies of pedestrian position prediction, this paper explores the combination of 2D background stripping modeling and spatiotemporal sequence. More specifically, a pedestrian position prediction method based on AlphaPose human pose estimation and PredRNN++ is proposed, and a data preprocessing method which can improve the test accuracy is also proposed. The validity of the proposed method is experimentally shown. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Autonomous driving / Pedestrians Position Prediction / Human Pose Estimation / Spatiotemporal Prediction |
| Paper # | PRMU2021-12 |
| Date of Issue | 2021-08-19 (PRMU) |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 2021/8/26(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Online |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | CV/PR techniques for human-robot cooperation |
| Chair | Seiichi Uchida(Kyushu Univ.) |
| Vice Chair | Masakazu Iwamura(Osaka Pref. Univ.) / Mitsuru Anpai(Denso IT Lab.) |
| Secretary | Masakazu Iwamura(NTT) / Mitsuru Anpai(Tottori Univ.) |
| Assistant | Kouta Yamaguchi(CyberAgent) / Yusuke Matsui(Univ. of Tokyo) |
| Paper Information | |
| Registration To | Technical Committee on Pattern Recognition and Media Understanding |
|---|---|
| Language | ENG-JTITLE |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Pedestrian Position Estimation by Integrating AlphaPose Based Human Pose Estimation and PredRNN++ Based Spatiotemporal Series Prediction for Autonomous Driving Systems |
| Sub Title (in English) | |
| Keyword(1) | Autonomous driving |
| Keyword(2) | Pedestrians Position Prediction |
| Keyword(3) | Human Pose Estimation |
| Keyword(4) | Spatiotemporal Prediction |
| 1st Author's Name | Zhaowei Zhang |
| 1st Author's Affiliation | Waseda Univ.(Waseda U.) |
| 2nd Author's Name | Eiji Fukuzawa |
| 2nd Author's Affiliation | Waseda Univ./Yazaki Corporation(Waseda U./Yazaki Corp.) |
| 3rd Author's Name | Jun Ohya |
| 3rd Author's Affiliation | Waseda Univ.(Waseda U.) |
| Date | 2021-08-26 |
| Paper # | PRMU2021-12 |
| Volume (vol) | vol.121 |
| Number (no) | PRMU-155 |
| Page | pp.pp.30-35(PRMU), |
| #Pages | 6 |
| Date of Issue | 2021-08-19 (PRMU) |