InsFusion: Rethink Instance-level LiDAR-Camera Fusion for 3D Object Detection
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912711858716672 |
|---|---|
| author | Xia, Zhongyu Yang, Hansong Wang, Yongtao |
| author_facet | Xia, Zhongyu Yang, Hansong Wang, Yongtao |
| contents | Three-dimensional Object Detection from multi-view cameras and LiDAR is a crucial component for autonomous driving and smart transportation. However, in the process of basic feature extraction, perspective transformation, and feature fusion, noise and error will gradually accumulate. To address this issue, we propose InsFusion, which can extract proposals from both raw and fused features and utilizes these proposals to query the raw features, thereby mitigating the impact of accumulated errors. Additionally, by incorporating attention mechanisms applied to the raw features, it thereby mitigates the impact of accumulated errors. Experiments on the nuScenes dataset demonstrate that InsFusion is compatible with various advanced baseline methods and delivers new state-of-the-art performance for 3D object detection. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_08374 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | InsFusion: Rethink Instance-level LiDAR-Camera Fusion for 3D Object Detection Xia, Zhongyu Yang, Hansong Wang, Yongtao Computer Vision and Pattern Recognition Three-dimensional Object Detection from multi-view cameras and LiDAR is a crucial component for autonomous driving and smart transportation. However, in the process of basic feature extraction, perspective transformation, and feature fusion, noise and error will gradually accumulate. To address this issue, we propose InsFusion, which can extract proposals from both raw and fused features and utilizes these proposals to query the raw features, thereby mitigating the impact of accumulated errors. Additionally, by incorporating attention mechanisms applied to the raw features, it thereby mitigates the impact of accumulated errors. Experiments on the nuScenes dataset demonstrate that InsFusion is compatible with various advanced baseline methods and delivers new state-of-the-art performance for 3D object detection. |
| title | InsFusion: Rethink Instance-level LiDAR-Camera Fusion for 3D Object Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2509.08374 |