Research map / 3D vision / 3D generation
MV3DIS: Multi-View Mask Matching via 3D Guides for Zero-Shot 3D Instance Segmentation
Yibo Zhao, Yigong Zhang, Jin Xie
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 region refinement (ablation) | coarse 3D segmentation baseline · ScanNetV2 class-agnostic | mAP: 33.3 -> 35.5 | Improved |
| add 3D-guided mask matching (ablation) | baseline + RR · ScanNetV2 class-agnostic | mAP: 35.5 -> 37.0 | Improved |
| add depth consistency weight (ablation) | baseline + RR + 3DG-MM · ScanNetV2 class-agnostic | mAP: 37.0 -> 38.5 | Improved |
| add 3D-guided mask matching (ablation) | baseline + RR · ScanNetV2 class-agnostic | AP50: 53.2 -> 57.6 | Improved |
| add all three components (ablation) | coarse baseline · ScanNetV2 class-agnostic | mAP: 33.3 -> 38.5 | Improved |
| depth tolerance alpha 0.05 instead of 0.01 (ablation) | MV3DIS · ScanNetV2 class-agnostic | mAP: 37.4 -> 38.5 | Improved |
| depth tolerance alpha 0.05 instead of 0.20 (ablation) | MV3DIS · ScanNetV2 class-agnostic | mAP: 36.0 -> 38.5 | Improved |
| merging threshold 0.5 instead of 0.3 (ablation) | MV3DIS · ScanNetV2 class-agnostic | mAP: 33.4 -> 38.5 | Improved |
| merging threshold 0.5 instead of 0.9 (ablation) | MV3DIS · ScanNetV2 class-agnostic | mAP: 31.7 -> 38.5 | Improved |
| use MV3DIS instead of SAM2Object | zero-shot 3D instance segmentation · ScanNetV2 class-agnostic | mAP: 34.0 -> 38.5 | Improved |
| zero-shot MV3DIS vs supervised Mask3D | ScanNet200 class-agnostic · ScanNet200 class-agnostic | AP50: 51.2 -> 54.7 | Improved |
| zero-shot MV3DIS vs Mask3D trained on ScanNetV2 | ScanNet++ class-agnostic · ScanNet++ class-agnostic | mAP: 9.9 -> 22.0 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- SAM2Object — SAM2Object: Consolidating View Consistency via SAM2 for Zero-Shot 3D Instance Segmentation (CVPR 2025) · better on 5 of 5 shared results
- SGS-3D — SGS-3D: High-Fidelity 3D Instance Segmentation via Reliable Semantic Mask Splitting and Growing (AAAI 2026) · 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
- Any3DIS: Class-Agnostic 3D Instance Segmentation by 2D Mask Tracking (CVPR 2025 · similarity 0.93)
- SAM2Object: Consolidating View Consistency via SAM2 for Zero-Shot 3D Instance Segmentation (CVPR 2025 · similarity 0.91)
- OnlineAnySeg: Online Zero-Shot 3D Segmentation by Visual Foundation Model Guided 2D Mask Merging (CVPR 2025 · similarity 0.90)
- MaskClustering: View Consensus based Mask Graph Clustering for Open-Vocabulary 3D Instance Segmentation (CVPR 2024 · similarity 0.90)
- Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidance (CVPR 2024 · similarity 0.89)
- SAI3D: Segment Any Instance in 3D Scenes (CVPR 2024 · similarity 0.89)
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.
