Research map / Detection and segmentation / 3D detection and driving
RobusTor3D: Robust Multimodal 3D Object Detector for Autonomous Driving by Vision-Language Knowledge Blending
Ying Yang, Hui Yin, Aixin Chong, Hui Wang, Zhengyin Liang
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 OTSG text descriptions to V-LKE (ablation) | RobusTor3D · KITTI val Pedestrian | 3D mAP: 63.49 -> 65.74 | Improved |
| add OTSG text descriptions to V-LKE (ablation) | RobusTor3D · KITTI val Car | 3D mAP: 87.63 -> 87.76 | Improved |
| add V-LKE image and CLIP vision encoder (ablation) | Voxel RCNN+ · KITTI val Pedestrian | 3D mAP: 60.34 -> 63.49 | Improved |
| add V-LKE image and CLIP vision encoder (ablation) | Voxel RCNN+ · KITTI val Car | 3D mAP: 86.78 -> 87.63 | Improved |
| add V-LKE image and CLIP vision encoder (ablation) | Voxel RCNN+ · KITTI val Cyclist | 3D mAP: 78.47 -> 79.43 | Improved |
| add ITCCL (ablation) | RobusTor3D · KITTI val Pedestrian | 3D mAP: 65.74 -> 66.48 | Improved |
| add IVCCL (ablation) | RobusTor3D · KITTI val Pedestrian | 3D mAP: 66.48 -> 66.91 | Improved |
| add IVCCL (ablation) | RobusTor3D · KITTI val Cyclist | 3D mAP: 79.61 -> 79.40 | No clear effect |
| freeze text encoder instead of image encoder (ablation) | RobusTor3D CLIP · KITTI val | 3D mAP: 76.3 -> 78.2 | Improved |
| add fusion of image and text features during training (ablation) | RobusTor3D · KITTI val modality drop (Car) | 3D mAP | Mixed |
| add RobusTor3D vision-language blending (ablation) | Voxel RCNN+ · KITTI val | 3D mAP: +1.50 +6.57 +0.93 | Improved |
| use RobusTor3D instead of SSLFusion | KITTI-C sensor misalignment · KITTI-C val | 3D mAP: 48.68 -> 59.14 | Improved |
Closest papers by meaning
- CorrBEV: Multi-View 3D Object Detection by Correlation Learning with Multi-modal Prototypes (CVPR 2025 · similarity 0.88)
- An Empirical Study of the Generalization Ability of Lidar 3D Object Detectors to Unseen Domains (CVPR 2024 · similarity 0.88)
- VLM2Scene: Self-Supervised Image-Text-LiDAR Learning with Foundation Models for Autonomous Driving Scene Understanding (AAAI 2024 · similarity 0.87)
- OpenAD: Open-World Autonomous Driving Benchmark for 3D Object Detection (NeurIPS 2025 · similarity 0.87)
- DriveGEN: Generalized and Robust 3D Detection in Driving via Controllable Text-to-Image Diffusion Generation (CVPR 2025 · similarity 0.87)
- ImOV3D: Learning Open Vocabulary Point Clouds 3D Object Detection from Only 2D Images (NeurIPS 2024 · similarity 0.86)
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.
