Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2504.09086 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913789745561600 |
|---|---|
| author | Long, Yunfei Kumar, Abhinav Liu, Xiaoming Morris, Daniel |
| author_facet | Long, Yunfei Kumar, Abhinav Liu, Xiaoming Morris, Daniel |
| contents | Radar hits reflect from points on both the boundary and internal to object outlines. This results in a complex distribution of radar hits that depends on factors including object category, size, and orientation. Current radar-camera fusion methods implicitly account for this with a black-box neural network. In this paper, we explicitly utilize a radar hit distribution model to assist fusion. First, we build a model to predict radar hit distributions conditioned on object properties obtained from a monocular detector. Second, we use the predicted distribution as a kernel to match actual measured radar points in the neighborhood of the monocular detections, generating matching scores at nearby positions. Finally, a fusion stage combines context with the kernel detector to refine the matching scores. Our method achieves the state-of-the-art radar-camera detection performance on nuScenes. Our source code is available at https://github.com/longyunf/riccardo. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_09086 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RICCARDO: Radar Hit Prediction and Convolution for Camera-Radar 3D Object Detection Long, Yunfei Kumar, Abhinav Liu, Xiaoming Morris, Daniel Computer Vision and Pattern Recognition Radar hits reflect from points on both the boundary and internal to object outlines. This results in a complex distribution of radar hits that depends on factors including object category, size, and orientation. Current radar-camera fusion methods implicitly account for this with a black-box neural network. In this paper, we explicitly utilize a radar hit distribution model to assist fusion. First, we build a model to predict radar hit distributions conditioned on object properties obtained from a monocular detector. Second, we use the predicted distribution as a kernel to match actual measured radar points in the neighborhood of the monocular detections, generating matching scores at nearby positions. Finally, a fusion stage combines context with the kernel detector to refine the matching scores. Our method achieves the state-of-the-art radar-camera detection performance on nuScenes. Our source code is available at https://github.com/longyunf/riccardo. |
| title | RICCARDO: Radar Hit Prediction and Convolution for Camera-Radar 3D Object Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2504.09086 |