KD360-VoxelBEV: LiDAR and 360-degree Camera Cross Modality Knowledge Distillation for Bird's-Eye-View Segmentation
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_ | 1866915681742618624 |
|---|---|
| author | E, Wenke Sun, Yixin Liu, Jiaxu Shum, Hubert P. H. Atapour-Abarghouei, Amir Breckon, Toby P. |
| author_facet | E, Wenke Sun, Yixin Liu, Jiaxu Shum, Hubert P. H. Atapour-Abarghouei, Amir Breckon, Toby P. |
| contents | We present the first cross-modality distillation framework specifically tailored for single-panoramic-camera Bird's-Eye-View (BEV) segmentation. Our approach leverages a novel LiDAR image representation fused from range, intensity and ambient channels, together with a voxel-aligned view transformer that preserves spatial fidelity while enabling efficient BEV processing. During training, a high-capacity LiDAR and camera fusion Teacher network extracts both rich spatial and semantic features for cross-modality knowledge distillation into a lightweight Student network that relies solely on a single 360-degree panoramic camera image. Extensive experiments on the Dur360BEV dataset demonstrate that our teacher model significantly outperforms existing camera-based BEV segmentation methods, achieving a 25.6\% IoU improvement. Meanwhile, the distilled Student network attains competitive performance with an 8.5\% IoU gain and state-of-the-art inference speed of 31.2 FPS. Moreover, evaluations on KITTI-360 (two fisheye cameras) confirm that our distillation framework generalises to diverse camera setups, underscoring its feasibility and robustness. This approach reduces sensor complexity and deployment costs while providing a practical solution for efficient, low-cost BEV segmentation in real-world autonomous driving. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_15311 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | KD360-VoxelBEV: LiDAR and 360-degree Camera Cross Modality Knowledge Distillation for Bird's-Eye-View Segmentation E, Wenke Sun, Yixin Liu, Jiaxu Shum, Hubert P. H. Atapour-Abarghouei, Amir Breckon, Toby P. Computer Vision and Pattern Recognition We present the first cross-modality distillation framework specifically tailored for single-panoramic-camera Bird's-Eye-View (BEV) segmentation. Our approach leverages a novel LiDAR image representation fused from range, intensity and ambient channels, together with a voxel-aligned view transformer that preserves spatial fidelity while enabling efficient BEV processing. During training, a high-capacity LiDAR and camera fusion Teacher network extracts both rich spatial and semantic features for cross-modality knowledge distillation into a lightweight Student network that relies solely on a single 360-degree panoramic camera image. Extensive experiments on the Dur360BEV dataset demonstrate that our teacher model significantly outperforms existing camera-based BEV segmentation methods, achieving a 25.6\% IoU improvement. Meanwhile, the distilled Student network attains competitive performance with an 8.5\% IoU gain and state-of-the-art inference speed of 31.2 FPS. Moreover, evaluations on KITTI-360 (two fisheye cameras) confirm that our distillation framework generalises to diverse camera setups, underscoring its feasibility and robustness. This approach reduces sensor complexity and deployment costs while providing a practical solution for efficient, low-cost BEV segmentation in real-world autonomous driving. |
| title | KD360-VoxelBEV: LiDAR and 360-degree Camera Cross Modality Knowledge Distillation for Bird's-Eye-View Segmentation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.15311 |