Research map / Detection and segmentation / 3D detection and driving
DSERT-RoLL: Robust Multi-Modal Perception for Diverse Driving Conditions with Stereo Event-RGB-Thermal Cameras, 4D Radar, and Dual-LiDAR
Hoonhee Cho, Jae-Young Kang, Yuhwan Jeong, Yunseo Yang, Wonyoung Lee, Youngho Kim, Kuk-Jin Yoon
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add 4D radar to LiDAR (ablation) | Ours · DSERT-RoLL 3D (Heavy Snow) | AP@0.5 (3D): 54.14 -> 69.96 | Improved |
| add RGB to LiDAR (ablation) | Ours · DSERT-RoLL 3D (Heavy Snow) | AP@0.5 (3D): 54.14 -> 59.43 | Improved |
| add RGB to LiDAR + 4D radar (ablation) | Ours · DSERT-RoLL 3D (Heavy Rain) | AP@0.5 (3D): 67.41 -> 79.03 | Improved |
| add RGB to LiDAR + 4D radar (ablation) | Ours · DSERT-RoLL 3D (Fog) | AP@0.5 (3D): 67.41 -> 67.38 | No clear effect |
| add event camera to R+4R+L (ablation) | Ours · DSERT-RoLL 3D (Fog) | AP@0.5 (3D): 67.38 -> 71.45 | Improved |
| add thermal camera to R+4R+L (ablation) | Ours · DSERT-RoLL 3D (Fog) | AP@0.5 (3D): 67.38 -> 71.00 | Improved |
| add event and thermal to R+4R+L (ablation) | Ours · DSERT-RoLL 3D (Clear) | AP@0.5 (3D): 88.35 -> 90.30 | Improved |
| add event to R+4R+L (ablation) | Ours · DSERT-RoLL 3D (Overexposed) | AP@0.5 (3D): 80.34 -> 83.37 | Improved |
| add event and thermal to R+4R+L (ablation) | Ours · DSERT-RoLL 3D (HDR) | AP@0.5 (3D): 83.93 -> 86.33 | Improved |
| use Ours instead of SAMFusion | Fusion detector · DSERT-RoLL 3D (Fog) | AP@0.5 (3D): 65.13 -> 71.42 | Improved |
| use multi-modal GM-DETR R+T+E instead of RGB-only DEIM | 2D detector · DSERT-RoLL 2D (Clear) | AP@0.5 (2D): 81.85 -> 90.36 | Improved |
| use thermal instead of RGB input with stereo DSGN | DSGN · DSERT-RoLL 3D (Light Snow) | AP@0.5 (3D): 25.94 -> 36.52 | Improved |
Closest papers by meaning
- V2X-Radar: A Multi-modal Dataset with 4D Radar for Cooperative Perception (NeurIPS 2025 · similarity 0.89)
- Multi-Modal 3D Object Detection in Autonomous Driving: A Survey (IJCV 2023 · similarity 0.88)
- The Road Ahead in Autonomous Driving: The KITScenes Multimodal Dataset (NeurIPS 2026 · similarity 0.86)
- RadarOcc: Robust 3D Occupancy Prediction with 4D Imaging Radar (NeurIPS 2024 · similarity 0.86)
- TUMTraf V2X Cooperative Perception Dataset (CVPR 2024 · similarity 0.85)
- L4DR: LiDAR-4DRadar Fusion for Weather-Robust 3D Object Detection (AAAI 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
