MambaFusion: Adaptive State-Space Fusion for Multimodal 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_ | 1866917269154562048 |
|---|---|
| author | Narayanan, Venkatraman Sai, Bala Ahuja, Rahul Likhar, Pratik Kumar, Varun Ravi Yogamani, Senthil |
| author_facet | Narayanan, Venkatraman Sai, Bala Ahuja, Rahul Likhar, Pratik Kumar, Varun Ravi Yogamani, Senthil |
| contents | Reliable 3D object detection is fundamental to autonomous driving, and multimodal fusion algorithms using cameras and LiDAR remain a persistent challenge. Cameras provide dense visual cues but ill posed depth; LiDAR provides a precise 3D structure but sparse coverage. Existing BEV-based fusion frameworks have made good progress, but they have difficulties including inefficient context modeling, spatially invariant fusion, and reasoning under uncertainty. We introduce MambaFusion, a unified multi-modal detection framework that achieves efficient, adaptive, and physically grounded 3D perception. MambaFusion interleaves selective state-space models (SSMs) with windowed transformers to propagate the global context in linear time while preserving local geometric fidelity. A multi-modal token alignment (MTA) module and reliability-aware fusion gates dynamically re-weight camera-LiDAR features based on spatial confidence and calibration consistency. Finally, a structure-conditioned diffusion head integrates graph-based reasoning with uncertainty-aware denoising, enforcing physical plausibility, and calibrated confidence. MambaFusion establishes new state-of-the-art performance on nuScenes benchmarks while operating with linear-time complexity. The framework demonstrates that coupling SSM-based efficiency with reliability-driven fusion yields robust, temporally stable, and interpretable 3D perception for real-world autonomous driving systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_08126 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MambaFusion: Adaptive State-Space Fusion for Multimodal 3D Object Detection Narayanan, Venkatraman Sai, Bala Ahuja, Rahul Likhar, Pratik Kumar, Varun Ravi Yogamani, Senthil Computer Vision and Pattern Recognition Reliable 3D object detection is fundamental to autonomous driving, and multimodal fusion algorithms using cameras and LiDAR remain a persistent challenge. Cameras provide dense visual cues but ill posed depth; LiDAR provides a precise 3D structure but sparse coverage. Existing BEV-based fusion frameworks have made good progress, but they have difficulties including inefficient context modeling, spatially invariant fusion, and reasoning under uncertainty. We introduce MambaFusion, a unified multi-modal detection framework that achieves efficient, adaptive, and physically grounded 3D perception. MambaFusion interleaves selective state-space models (SSMs) with windowed transformers to propagate the global context in linear time while preserving local geometric fidelity. A multi-modal token alignment (MTA) module and reliability-aware fusion gates dynamically re-weight camera-LiDAR features based on spatial confidence and calibration consistency. Finally, a structure-conditioned diffusion head integrates graph-based reasoning with uncertainty-aware denoising, enforcing physical plausibility, and calibrated confidence. MambaFusion establishes new state-of-the-art performance on nuScenes benchmarks while operating with linear-time complexity. The framework demonstrates that coupling SSM-based efficiency with reliability-driven fusion yields robust, temporally stable, and interpretable 3D perception for real-world autonomous driving systems. |
| title | MambaFusion: Adaptive State-Space Fusion for Multimodal 3D Object Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.08126 |