RAPTR: Radar-based 3D Pose Estimation using Transformer
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909898416062464 |
|---|---|
| author | Kato, Sorachi Yataka, Ryoma Wang, Pu Perry Miraldo, Pedro Fujihashi, Takuya Boufounos, Petros |
| author_facet | Kato, Sorachi Yataka, Ryoma Wang, Pu Perry Miraldo, Pedro Fujihashi, Takuya Boufounos, Petros |
| contents | Radar-based indoor 3D human pose estimation typically relied on fine-grained 3D keypoint labels, which are costly to obtain especially in complex indoor settings involving clutter, occlusions, or multiple people. In this paper, we propose \textbf{RAPTR} (RAdar Pose esTimation using tRansformer) under weak supervision, using only 3D BBox and 2D keypoint labels which are considerably easier and more scalable to collect. Our RAPTR is characterized by a two-stage pose decoder architecture with a pseudo-3D deformable attention to enhance (pose/joint) queries with multi-view radar features: a pose decoder estimates initial 3D poses with a 3D template loss designed to utilize the 3D BBox labels and mitigate depth ambiguities; and a joint decoder refines the initial poses with 2D keypoint labels and a 3D gravity loss. Evaluated on two indoor radar datasets, RAPTR outperforms existing methods, reducing joint position error by $34.3\%$ on HIBER and $76.9\%$ on MMVR. Our implementation is available at https://github.com/merlresearch/radar-pose-transformer. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_08387 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RAPTR: Radar-based 3D Pose Estimation using Transformer Kato, Sorachi Yataka, Ryoma Wang, Pu Perry Miraldo, Pedro Fujihashi, Takuya Boufounos, Petros Computer Vision and Pattern Recognition Artificial Intelligence Signal Processing Radar-based indoor 3D human pose estimation typically relied on fine-grained 3D keypoint labels, which are costly to obtain especially in complex indoor settings involving clutter, occlusions, or multiple people. In this paper, we propose \textbf{RAPTR} (RAdar Pose esTimation using tRansformer) under weak supervision, using only 3D BBox and 2D keypoint labels which are considerably easier and more scalable to collect. Our RAPTR is characterized by a two-stage pose decoder architecture with a pseudo-3D deformable attention to enhance (pose/joint) queries with multi-view radar features: a pose decoder estimates initial 3D poses with a 3D template loss designed to utilize the 3D BBox labels and mitigate depth ambiguities; and a joint decoder refines the initial poses with 2D keypoint labels and a 3D gravity loss. Evaluated on two indoor radar datasets, RAPTR outperforms existing methods, reducing joint position error by $34.3\%$ on HIBER and $76.9\%$ on MMVR. Our implementation is available at https://github.com/merlresearch/radar-pose-transformer. |
| title | RAPTR: Radar-based 3D Pose Estimation using Transformer |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Signal Processing |
| url | https://arxiv.org/abs/2511.08387 |