Event-Based Pedestrian Detection Using Dynamic Vision Sensors
Pedestrian detection has attracted great research attention in video surveillance, traffic statistics, and especially in autonomous driving. To date, almost all pedestrian detection solutions are derived from conventional framed-based image sensors with limited reaction speed and high data redundanc...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-04-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/10/8/888 |
| _version_ | 1797538399158009856 |
|---|---|
| author | Jixiang Wan Ming Xia Zunkai Huang Li Tian Xiaoying Zheng Victor Chang Yongxin Zhu Hui Wang |
| author_facet | Jixiang Wan Ming Xia Zunkai Huang Li Tian Xiaoying Zheng Victor Chang Yongxin Zhu Hui Wang |
| author_sort | Jixiang Wan |
| collection | DOAJ |
| description | Pedestrian detection has attracted great research attention in video surveillance, traffic statistics, and especially in autonomous driving. To date, almost all pedestrian detection solutions are derived from conventional framed-based image sensors with limited reaction speed and high data redundancy. Dynamic vision sensor (DVS), which is inspired by biological retinas, efficiently captures the visual information with sparse, asynchronous events rather than dense, synchronous frames. It can eliminate redundant data transmission and avoid motion blur or data leakage in high-speed imaging applications. However, it is usually impractical to directly apply the event streams to conventional object detection algorithms. For this issue, we first propose a novel event-to-frame conversion method by integrating the inherent characteristics of events more efficiently. Moreover, we design an improved feature extraction network that can reuse intermediate features to further reduce the computational effort. We evaluate the performance of our proposed method on a custom dataset containing multiple real-world pedestrian scenes. The results indicate that our proposed method raised its pedestrian detection accuracy by about 5.6–10.8%, and its detection speed is nearly 20% faster than previously reported methods. Furthermore, it can achieve a processing speed of about 26 FPS and an AP of 87.43% when implanted on a single CPU so that it fully meets the requirement of real-time detection. |
| first_indexed | 2024-03-10T12:30:59Z |
| format | Article |
| id | doaj.art-7bcd7a3f49744cd29d7e14c40fa4155e |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T12:30:59Z |
| publishDate | 2021-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-7bcd7a3f49744cd29d7e14c40fa4155e2023-11-21T14:40:40ZengMDPI AGElectronics2079-92922021-04-0110888810.3390/electronics10080888Event-Based Pedestrian Detection Using Dynamic Vision SensorsJixiang Wan0Ming Xia1Zunkai Huang2Li Tian3Xiaoying Zheng4Victor Chang5Yongxin Zhu6Hui Wang7Shanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaShanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaShanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaShanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaShanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaSchool of Computing & Digital Technologies, Teesside University, Middlesbrough TS1 3JN, UKShanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaShanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, ChinaPedestrian detection has attracted great research attention in video surveillance, traffic statistics, and especially in autonomous driving. To date, almost all pedestrian detection solutions are derived from conventional framed-based image sensors with limited reaction speed and high data redundancy. Dynamic vision sensor (DVS), which is inspired by biological retinas, efficiently captures the visual information with sparse, asynchronous events rather than dense, synchronous frames. It can eliminate redundant data transmission and avoid motion blur or data leakage in high-speed imaging applications. However, it is usually impractical to directly apply the event streams to conventional object detection algorithms. For this issue, we first propose a novel event-to-frame conversion method by integrating the inherent characteristics of events more efficiently. Moreover, we design an improved feature extraction network that can reuse intermediate features to further reduce the computational effort. We evaluate the performance of our proposed method on a custom dataset containing multiple real-world pedestrian scenes. The results indicate that our proposed method raised its pedestrian detection accuracy by about 5.6–10.8%, and its detection speed is nearly 20% faster than previously reported methods. Furthermore, it can achieve a processing speed of about 26 FPS and an AP of 87.43% when implanted on a single CPU so that it fully meets the requirement of real-time detection.https://www.mdpi.com/2079-9292/10/8/888dynamic vision sensorevent datapedestrian detectionautonomous driving |
| spellingShingle | Jixiang Wan Ming Xia Zunkai Huang Li Tian Xiaoying Zheng Victor Chang Yongxin Zhu Hui Wang Event-Based Pedestrian Detection Using Dynamic Vision Sensors Electronics dynamic vision sensor event data pedestrian detection autonomous driving |
| title | Event-Based Pedestrian Detection Using Dynamic Vision Sensors |
| title_full | Event-Based Pedestrian Detection Using Dynamic Vision Sensors |
| title_fullStr | Event-Based Pedestrian Detection Using Dynamic Vision Sensors |
| title_full_unstemmed | Event-Based Pedestrian Detection Using Dynamic Vision Sensors |
| title_short | Event-Based Pedestrian Detection Using Dynamic Vision Sensors |
| title_sort | event based pedestrian detection using dynamic vision sensors |
| topic | dynamic vision sensor event data pedestrian detection autonomous driving |
| url | https://www.mdpi.com/2079-9292/10/8/888 |
| work_keys_str_mv | AT jixiangwan eventbasedpedestriandetectionusingdynamicvisionsensors AT mingxia eventbasedpedestriandetectionusingdynamicvisionsensors AT zunkaihuang eventbasedpedestriandetectionusingdynamicvisionsensors AT litian eventbasedpedestriandetectionusingdynamicvisionsensors AT xiaoyingzheng eventbasedpedestriandetectionusingdynamicvisionsensors AT victorchang eventbasedpedestriandetectionusingdynamicvisionsensors AT yongxinzhu eventbasedpedestriandetectionusingdynamicvisionsensors AT huiwang eventbasedpedestriandetectionusingdynamicvisionsensors |