Research map / 3D vision / 3D generation
MamV2XCalib: V2X-based Target-less Infrastructure Camera Calibration with State Space Model
Yaoye Zhu, Zhe Wang, Yan Wang
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 Mamba temporal fusion (vs Timesformer) (ablation) | MamV2XCalib · V2X-Seq (-20,+20 deg) | mean error (deg): 0.6683 -> 0.6313 | Improved |
| add Mamba temporal fusion (vs Timesformer) (ablation) | MamV2XCalib · V2X-Seq (-20,+20 deg) | std (deg): 0.3808 -> 0.3211 | Improved |
| add temporal module (ablation) | MamV2XCalib · V2X-Seq (-20,+20 deg) | std (deg): 2.0013 -> 0.3211 | Improved |
| add temporal module (ablation) | MamV2XCalib · V2X-Seq (-20,+20 deg) | mean error (deg): 0.9742 -> 0.6313 | Improved |
| add iterative refinement (10 iterations) (ablation) | MamV2XCalib · V2X-Seq (-20,+20 deg) | mean error (deg): 0.7036 -> 0.6313 | Improved |
| add iterative refinement (10 iterations) (ablation) | MamV2XCalib · V2X-Seq (-20,+20 deg) | std (deg): 1.6162 -> 0.3211 | Improved |
| replace 3D feature matching with multi-scale 4D correlation volume (ablation) | MamV2XCalib · V2X-Seq (-20,+20 deg) | mean error (deg): 1.0741 -> 0.9742 | Improved |
| replace 3D feature matching with multi-scale 4D correlation volume (ablation) | MamV2XCalib · V2X-Seq (-20,+20 deg) | parameters (M): 70 -> 41 | Improved |
| use Mamba fusion instead of least square fitting of per-frame estimates (ablation) | MamV2XCalib · V2X-Seq (-20,+20 deg) | mean error (deg): 0.820 -> 0.631 | Improved |
| use Mamba fusion instead of stacking multi-frame points (ablation) | MamV2XCalib · V2X-Seq (-20,+20 deg) | mean error (deg): 0.952 -> 0.631 | Improved |
| use MamV2XCalib instead of LCCNet | MamV2XCalib · V2X-Seq (-20,+20 deg) | std (deg): 2.5336 -> 0.3211 | Improved |
| use MamV2XCalib instead of LCCNet | MamV2XCalib · TUMTraf-V2X (-5,+5 deg) | mean error (deg): 0.389 -> 0.267 | Improved |
Closest papers by meaning
- UniCal: Unified Neural Sensor Calibration (ECCV 2024 · similarity 0.87)
- DI-V2X: Learning Domain-Invariant Representation for Vehicle-Infrastructure Collaborative 3D Object Detection (AAAI 2024 · similarity 0.84)
- V2X-Radar: A Multi-modal Dataset with 4D Radar for Cooperative Perception (NeurIPS 2025 · similarity 0.84)
- UrbanIng-V2X: A Large-Scale Multi-Vehicle, Multi-Infrastructure Dataset Across Multiple Intersections for Cooperative Perception (NeurIPS 2025 · similarity 0.84)
- HoloVIC: Large-scale Dataset and Benchmark for Multi-Sensor Holographic Intersection and Vehicle-Infrastructure Cooperative (CVPR 2024 · similarity 0.83)
- TUMTraf V2X Cooperative Perception Dataset (CVPR 2024 · similarity 0.83)
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.
