Research map / 3D vision / 3D generation
PanSt3R: Multi-view Consistent Panoptic Segmentation
Lojze Zust, Yohann Cabon, Juliette Marrie, Leonid Antsfeld, Boris Chidlovskii, Jerome Revaud, Gabriela Csurka
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 DINOv2 2D features to MUSt3R 3D features (ablation) | PanSt3R · ScanNet++ val (224, +LUDVIG) | PQ: 46.4 -> 50.4 | Improved |
| add MUSt3R 3D features to DINOv2 2D features (ablation) | PanSt3R · ScanNet++ val (224, +LUDVIG) | PQ: 35.7 -> 50.4 | Improved |
| add QUBO mask merging (ablation) | PanSt3R + LUDVIG (no reg) · Hypersim | PQ: 58.1 -> 66.7 | Improved |
| add QUBO mask merging (ablation) | PanSt3R + LUDVIG (no reg) · ScanNet | PQ: 60.2 -> 67.3 | Improved |
| add QUBO mask merging (ablation) | PanSt3R + LUDVIG (no reg) · Replica | PQ: 60.8 -> 60.7 | No clear effect |
| add QUBO mask merging to direct prediction (ablation) | PanSt3R (rendered views) · Hypersim | PQ: 51.6 -> 56.5 | Improved |
| add QUBO mask merging to direct prediction (ablation) | PanSt3R (rendered views) · ScanNet | PQ: 59.5 -> 65.7 | Improved |
| add panoptic regularization to 3DGS (QUBO off) (ablation) | PanSt3R + LUDVIG · ScanNet++ val | PQ: 50.8 -> 55.2 | Improved |
| add panoptic regularization to 3DGS (QUBO on) (ablation) | PanSt3R + LUDVIG · ScanNet++ val | PQ: 52.0 -> 54.7 | Hurt |
| add panoptic regularization to 3DGS (QUBO on) (ablation) | PanSt3R + LUDVIG · Hypersim | PQ: 66.7 -> 66.3 | Hurt |
| use 50 keyframes instead of 10 (ablation) | PanSt3R · ScanNet++ seen set | PQ: 49.3 -> 58.2 | Improved |
| use 100 keyframes instead of 50 (ablation) | PanSt3R · ScanNet++ seen set | PQ: 58.2 -> 58.2 | No clear effect |
Later papers that compared with this method
2 later accepted papers report PanSt3R as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- SegVGGT: Joint 3D Reconstruction and Instance Segmentation from Multi-View Images (ECCV 2026 · better on 4 of 4 shared results)
- Pano3D: Unified 3D Reconstruction and Panoptic Segmentation (ECCV 2026 · better on 3 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DM-NeRF — DM-NeRF: 3D Scene Geometry Decomposition and Manipulation from 2D Images (ICLR 2023) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- MC-PanDA++: Simpler, Stronger, and More Robust Domain-Adaptive Panoptic Segmentation (IJCV 2026 · similarity 0.90)
- Real-Time Panoptic Segmentation From Dense Detections (CVPR 2020 · similarity 0.86)
- Multi-view Consistent 3D Panoptic Scene Understanding (AAAI 2025 · similarity 0.86)
- Scene-Centric Unsupervised Panoptic Segmentation (CVPR 2025 · similarity 0.85)
- How Do Images Align and Complement LiDAR? Towards a Harmonized Multi-modal 3D Panoptic Segmentation (ICML 2025 · similarity 0.84)
- PanoRecon: Real-Time Panoptic 3D Reconstruction from Monocular Video (CVPR 2024 · 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.
