Vulnerable Road User Skeletal Pose Estimation Using mmWave Radars
A skeletal pose estimation method, named RVRU-Pose, is proposed to estimate the skeletal pose of vulnerable road users based on distributed non-coherent mmWave radar. In view of the limitation that existing methods for skeletal pose estimation are only applicable to small scenes, this paper proposes...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-02-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/16/4/633 |
| _version_ | 1797297072674701312 |
|---|---|
| author | Zhiyuan Zeng Xingdong Liang Yanlei Li Xiangwei Dang |
| author_facet | Zhiyuan Zeng Xingdong Liang Yanlei Li Xiangwei Dang |
| author_sort | Zhiyuan Zeng |
| collection | DOAJ |
| description | A skeletal pose estimation method, named RVRU-Pose, is proposed to estimate the skeletal pose of vulnerable road users based on distributed non-coherent mmWave radar. In view of the limitation that existing methods for skeletal pose estimation are only applicable to small scenes, this paper proposes a strategy that combines radar intensity heatmaps and coordinate heatmaps as input to a deep learning network. In addition, we design a multi-resolution data augmentation and training method suitable for radar to achieve target pose estimation for remote and multi-target application scenarios. Experimental results show that RVRU-Pose can achieve better than 2 cm average localization accuracy for different subjects in different scenarios, which is superior in terms of accuracy and time compared to existing state-of-the-art methods for human skeletal pose estimation with radar. As an essential performance parameter of radar, the impact of angular resolution on the estimation accuracy of a skeletal pose is quantitatively analyzed and evaluated in this paper. Finally, RVRU-Pose has also been extended to the task of estimating the skeletal pose of a cyclist, reflecting the strong scalability of the proposed method. |
| first_indexed | 2024-03-07T22:15:57Z |
| format | Article |
| id | doaj.art-65d5f173b42c41aeb8cd2244392a92cd |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-07T22:15:57Z |
| publishDate | 2024-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-65d5f173b42c41aeb8cd2244392a92cd2024-02-23T15:32:56ZengMDPI AGRemote Sensing2072-42922024-02-0116463310.3390/rs16040633Vulnerable Road User Skeletal Pose Estimation Using mmWave RadarsZhiyuan Zeng0Xingdong Liang1Yanlei Li2Xiangwei Dang3National Key Laboratory of Microwave Imaging Technology, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, ChinaNational Key Laboratory of Microwave Imaging Technology, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, ChinaNational Key Laboratory of Microwave Imaging Technology, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, ChinaNational Key Laboratory of Microwave Imaging Technology, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, ChinaA skeletal pose estimation method, named RVRU-Pose, is proposed to estimate the skeletal pose of vulnerable road users based on distributed non-coherent mmWave radar. In view of the limitation that existing methods for skeletal pose estimation are only applicable to small scenes, this paper proposes a strategy that combines radar intensity heatmaps and coordinate heatmaps as input to a deep learning network. In addition, we design a multi-resolution data augmentation and training method suitable for radar to achieve target pose estimation for remote and multi-target application scenarios. Experimental results show that RVRU-Pose can achieve better than 2 cm average localization accuracy for different subjects in different scenarios, which is superior in terms of accuracy and time compared to existing state-of-the-art methods for human skeletal pose estimation with radar. As an essential performance parameter of radar, the impact of angular resolution on the estimation accuracy of a skeletal pose is quantitatively analyzed and evaluated in this paper. Finally, RVRU-Pose has also been extended to the task of estimating the skeletal pose of a cyclist, reflecting the strong scalability of the proposed method.https://www.mdpi.com/2072-4292/16/4/633mmWave radarskeletal pose estimationradar signal processingconvolutional neural network |
| spellingShingle | Zhiyuan Zeng Xingdong Liang Yanlei Li Xiangwei Dang Vulnerable Road User Skeletal Pose Estimation Using mmWave Radars Remote Sensing mmWave radar skeletal pose estimation radar signal processing convolutional neural network |
| title | Vulnerable Road User Skeletal Pose Estimation Using mmWave Radars |
| title_full | Vulnerable Road User Skeletal Pose Estimation Using mmWave Radars |
| title_fullStr | Vulnerable Road User Skeletal Pose Estimation Using mmWave Radars |
| title_full_unstemmed | Vulnerable Road User Skeletal Pose Estimation Using mmWave Radars |
| title_short | Vulnerable Road User Skeletal Pose Estimation Using mmWave Radars |
| title_sort | vulnerable road user skeletal pose estimation using mmwave radars |
| topic | mmWave radar skeletal pose estimation radar signal processing convolutional neural network |
| url | https://www.mdpi.com/2072-4292/16/4/633 |
| work_keys_str_mv | AT zhiyuanzeng vulnerableroaduserskeletalposeestimationusingmmwaveradars AT xingdongliang vulnerableroaduserskeletalposeestimationusingmmwaveradars AT yanleili vulnerableroaduserskeletalposeestimationusingmmwaveradars AT xiangweidang vulnerableroaduserskeletalposeestimationusingmmwaveradars |