| Presentation | 2021-10-22 A saliency estimation model for drivers' egocentric vision movies considering self-motion velocity Yuya Homma, Masashi Fujita, Takeshi Kohama, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In order to predict where a driver’s attention should be directed during driving, Kodama et al. have developed a saliency estimation model for self-motion images that take into account the receptive field characteristics of the early visual system and higher-order motion selection mechanisms. However, Kodama et al.’s model has some problems, such as reduced prediction accuracy when driving at low speeds, and the response of the receptive field model to extract features in dark areas remains an issue. In this study, to construct a mathematical model for a more accurate prediction of driver’s gaze while driving a car, we added the process of dynamically changing the frame-interval when calculating motion vectors and modified the on-center and off-center antagonistic receptive field responses to the Kodama et al.’s model. The results suggest that the moving-NSS (Moving-Normalized Scanpath Saliency) score, which indicates the prediction accuracy of the attentional area, improved for all subjects in the scenes with low-speed driving and the scenes with many dark features. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Driver’s egocentric vision / Attention prediction / Saliency map / Motion detection / Early visual system receptive fields |
| Paper # | HIP2021-44 |
| Date of Issue | 2021-10-14 (HIP) |
| Conference Information | |
| Committee | HIP |
|---|---|
| Conference Date | 2021/10/21(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Online |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Shuichi Sakamoto(Tohoku Univ.) |
| Vice Chair | Yuji Wada(Ritsumeikan Univ.) / Sachiko Kiyokawa(Nagoya Univ.) |
| Secretary | Yuji Wada(NTT) / Sachiko Kiyokawa(NICT) |
| Assistant | Yuki Yamada(Kyushu Univ.) / Daisuke Tanaka(Tottori Univ.) / Ippei Negishi(Kanazawa Inst. of Tech.) |
| Paper Information | |
| Registration To | Technical Committee on Human Information Processing |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | A saliency estimation model for drivers' egocentric vision movies considering self-motion velocity |
| Sub Title (in English) | |
| Keyword(1) | Driver’s egocentric vision |
| Keyword(2) | Attention prediction |
| Keyword(3) | Saliency map |
| Keyword(4) | Motion detection |
| Keyword(5) | Early visual system receptive fields |
| 1st Author's Name | Yuya Homma |
| 1st Author's Affiliation | Kindai University(Kindai Univ.) |
| 2nd Author's Name | Masashi Fujita |
| 2nd Author's Affiliation | Kindai University(Kindai Univ.) |
| 3rd Author's Name | Takeshi Kohama |
| 3rd Author's Affiliation | Kindai University(Kindai Univ.) |
| Date | 2021-10-22 |
| Paper # | HIP2021-44 |
| Volume (vol) | vol.121 |
| Number (no) | HIP-211 |
| Page | pp.pp.75-80(HIP), |
| #Pages | 6 |
| Date of Issue | 2021-10-14 (HIP) |