Research map / Point clouds / Point cloud classification
UniFlow: Zero-Shot LiDAR Scene Flow for Autonomous Driving
Siyi Li, Qingwen Zhang, Ishan Khatri, Kyle Vedder, Eric Eaton, Deva Ramanan, Neehar Peri
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 |
|---|---|---|---|
| train on unified multi-dataset mixture (AV2, Waymo, nuScenes) instead of Waymo only (ablation) | Flow4D · TruckScenes zero-shot | Dynamic Mean EPE: 0.336 -> 0.310 | Improved |
| train on unified multi-dataset mixture instead of Waymo only (ablation) | SSF · TruckScenes zero-shot | Dynamic Mean EPE: 0.737 -> 0.577 | Improved |
| train on unified multi-dataset mixture instead of Waymo only (ablation) | DeltaFlow · TruckScenes zero-shot | Dynamic Mean EPE: 0.357 -> 0.331 | Improved |
| add height randomization and LiDAR ray dropping augmentation (ablation) | DeltaFlow (UniFlow) · TruckScenes zero-shot | Dynamic Mean EPE: 0.331 -> 0.283 | Improved |
| add augmentation to unified training (ablation) | SSF (UniFlow) · TruckScenes zero-shot | Dynamic Mean EPE: 0.577 -> 0.435 | Improved |
| add augmentation to unified training (ablation) | Flow4D (UniFlow) · TruckScenes zero-shot | Dynamic Mean EPE: 0.310 -> 0.301 | Improved |
| scale up Flow4D backbone to XL (ablation) | Flow4D (UniFlow) · TruckScenes zero-shot | Dynamic Mean EPE: 0.301 -> 0.281 | Improved |
| multi-dataset UniFlow training instead of dataset-specific training | Flow4D-XL · Argoverse 2 test | Dynamic Mean EPE: 0.145 -> 0.132 | Improved |
| multi-dataset UniFlow training instead of dataset-specific training | SSF · nuScenes val | Dynamic Mean EPE: 0.220 -> 0.144 | Improved |
| multi-dataset UniFlow training instead of dataset-specific training | DeltaFlow · Argoverse 2 test | Dynamic Mean EPE: 0.113 -> 0.118 | No clear effect |
| multi-dataset UniFlow training instead of best single-dataset model | DeltaFlow · AEVAScenes zero-shot | Dynamic Mean EPE: 0.391 -> 0.303 | Improved |
| multi-dataset UniFlow training instead of AV2-only for long range | DeltaFlow-LR · AV2 long-range | Dynamic Mean EPE: 0.259 -> 0.226 | Improved |
Closest papers by meaning
- ICP-Flow: LiDAR Scene Flow Estimation with ICP (CVPR 2024 · similarity 0.92)
- ZeroFlow: Scalable Scene Flow via Distillation (ICLR 2024 · similarity 0.89)
- Motion Cues from Image-based Point Tracking for LiDAR Scene Flow Estimation (NeurIPS 2026 · similarity 0.88)
- 3DSFLabelling: Boosting 3D Scene Flow Estimation by Pseudo Auto-labelling (CVPR 2024 · similarity 0.88)
- HiLoTs: High-Low Temporal Sensitive Representation Learning for Semi-Supervised LiDAR Segmentation in Autonomous Driving (CVPR 2025 · similarity 0.88)
- Zero-Shot Monocular Scene Flow Estimation in the Wild (CVPR 2025 · similarity 0.88)
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.
