Boosting Instance Awareness via Cross-View Correlation with 4D Radar and Camera for 3D Object Detection
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908850063409152 |
|---|---|
| author | Bai, Xiaokai Zheng, Lianqing Cao, Si-Yuan Zhang, Xiaohan Wu, Zhe Yu, Beinan Wang, Fang Bai, Jie Shen, Hui-Liang |
| author_facet | Bai, Xiaokai Zheng, Lianqing Cao, Si-Yuan Zhang, Xiaohan Wu, Zhe Yu, Beinan Wang, Fang Bai, Jie Shen, Hui-Liang |
| contents | 4D millimeter-wave radar has emerged as a promising sensing modality for autonomous driving due to its robustness and affordability. However, its sparse and weak geometric cues make reliable instance activation difficult, limiting the effectiveness of existing radar-camera fusion paradigms. BEV-level fusion offers global scene understanding but suffers from weak instance focus, while perspective-level fusion captures instance details but lacks holistic context. To address these limitations, we propose SIFormer, a scene-instance aware transformer for 3D object detection using 4D radar and camera. SIFormer first suppresses background noise during view transformation through segmentation- and depth-guided localization. It then introduces a cross-view activation mechanism that injects 2D instance cues into BEV space, enabling reliable instance awareness under weak radar geometry. Finally, a transformer-based fusion module aggregates complementary image semantics and radar geometry for robust perception. As a result, with the aim of enhancing instance awareness, SIFormer bridges the gap between the two paradigms, combining their complementary strengths to address inherent sparse nature of radar and improve detection accuracy. Experiments demonstrate that SIFormer achieves state-of-the-art performance on View-of-Delft, TJ4DRadSet and NuScenes datasets. Source code is available at github.com/shawnnnkb/SIFormer. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_20632 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Boosting Instance Awareness via Cross-View Correlation with 4D Radar and Camera for 3D Object Detection Bai, Xiaokai Zheng, Lianqing Cao, Si-Yuan Zhang, Xiaohan Wu, Zhe Yu, Beinan Wang, Fang Bai, Jie Shen, Hui-Liang Computer Vision and Pattern Recognition 4D millimeter-wave radar has emerged as a promising sensing modality for autonomous driving due to its robustness and affordability. However, its sparse and weak geometric cues make reliable instance activation difficult, limiting the effectiveness of existing radar-camera fusion paradigms. BEV-level fusion offers global scene understanding but suffers from weak instance focus, while perspective-level fusion captures instance details but lacks holistic context. To address these limitations, we propose SIFormer, a scene-instance aware transformer for 3D object detection using 4D radar and camera. SIFormer first suppresses background noise during view transformation through segmentation- and depth-guided localization. It then introduces a cross-view activation mechanism that injects 2D instance cues into BEV space, enabling reliable instance awareness under weak radar geometry. Finally, a transformer-based fusion module aggregates complementary image semantics and radar geometry for robust perception. As a result, with the aim of enhancing instance awareness, SIFormer bridges the gap between the two paradigms, combining their complementary strengths to address inherent sparse nature of radar and improve detection accuracy. Experiments demonstrate that SIFormer achieves state-of-the-art performance on View-of-Delft, TJ4DRadSet and NuScenes datasets. Source code is available at github.com/shawnnnkb/SIFormer. |
| title | Boosting Instance Awareness via Cross-View Correlation with 4D Radar and Camera for 3D Object Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.20632 |