Research map / 3D vision / 3D generation
Understanding Cross-Rig Generalization in Automotive Perception: a Multi-Rig Benchmark and Rig Variation Metrics
Tim Alexander Bader, Tim Eberhardt, Maximilian Dillitzer, Wilhelm Stork
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 FOV/overlap term (lambda_f) to RigCD (ablation) | RigCD BEVDet · diverse rigs | Spearman rho: 0.330 -> 0.718 | Improved |
| add FOV/overlap term (lambda_f) to RigCD (ablation) | RigCD BEVFusion · diverse rigs | Spearman rho: 0.331 -> 0.734 | Improved |
| add FOV/overlap term (lambda_f) to RigCD (ablation) | RigCD Fast-BEV · diverse rigs | Spearman rho: 0.679 -> 0.097 | Hurt |
| add FOV/overlap term (lambda_f) to RigCD (ablation) | RigCD PETR · diverse rigs | Spearman rho: 0.787 -> 0.804 | No clear effect |
| add translation term (lambda_t) to RigCD (ablation) | RigCD PETR · diverse rigs | Spearman rho: 0.220 -> 0.804 | Improved |
| add translation term (lambda_t) to RigCD (ablation) | RigCD BEVDet · diverse rigs | Spearman rho: 0.743 -> 0.718 | Hurt |
| add translation term (lambda_t) to RigCD (ablation) | RigCD BEVFusion · diverse rigs | Spearman rho: 0.719 -> 0.734 | No clear effect |
| add rotation term (lambda_r) to RigCD (ablation) | RigCD all models · diverse rigs | Spearman rho: 0.719 -> 0.718 | No clear effect |
| add camera-count penalty (alpha<1) to RigCD (ablation) | RigCD all models · diverse rigs | Spearman rho: 0.716 -> 0.718 | No clear effect |
| calibrate RigCD jointly across four models instead of separately | RigCD Fast-BEV · diverse rigs | Spearman rho: 0.097 -> 0.59 | Improved |
| calibrate RigCD jointly across four models instead of separately | RigCD BEVFusion · diverse rigs | Spearman rho: 0.734 -> 0.68 | Hurt |
| train BEVFusion on union of rig slices R1-R6 (ablation) | BEVFusion · unseen rigs R7, R8, R9 | relative mAP gap: -38.37%, -71.16%, -48.14% | Mixed |
Closest papers by meaning
- UniDrive: Towards Universal Driving Perception Across Camera Configurations (ICLR 2025 · similarity 0.86)
- Mind the Hitch: Dynamic Calibration and Articulated Perception for Autonomous Trucks (CVPR 2026 · similarity 0.86)
- Rethinking the Spatio-Temporal Alignment of End-to-End 3D Perception (AAAI 2026 · similarity 0.86)
- LiDAR Prompted Spatio-Temporal Multi-View Stereo for Autonomous Driving (CVPR 2026 · similarity 0.86)
- VGGDrive: Empowering Vision-Language Models with Cross-View Geometric Grounding for Autonomous Driving (CVPR 2026 · similarity 0.84)
- VR-Drive: Viewpoint-Robust End-to-End Driving with Feed-Forward 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.84)
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.
