AI Paper Atlas

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

CVPR 2026 · 3D vision › 3D generation · approach: Learning 3D reconstruction from images · 0 citations (OpenAlex)

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 changedWhereMeasuredEffect
add region refinement (ablation)coarse 3D segmentation baseline · ScanNetV2 class-agnosticmAP: 33.3 -> 35.5Improved
add 3D-guided mask matching (ablation)baseline + RR · ScanNetV2 class-agnosticmAP: 35.5 -> 37.0Improved
add depth consistency weight (ablation)baseline + RR + 3DG-MM · ScanNetV2 class-agnosticmAP: 37.0 -> 38.5Improved
add 3D-guided mask matching (ablation)baseline + RR · ScanNetV2 class-agnosticAP50: 53.2 -> 57.6Improved
add all three components (ablation)coarse baseline · ScanNetV2 class-agnosticmAP: 33.3 -> 38.5Improved
depth tolerance alpha 0.05 instead of 0.01 (ablation)MV3DIS · ScanNetV2 class-agnosticmAP: 37.4 -> 38.5Improved
depth tolerance alpha 0.05 instead of 0.20 (ablation)MV3DIS · ScanNetV2 class-agnosticmAP: 36.0 -> 38.5Improved
merging threshold 0.5 instead of 0.3 (ablation)MV3DIS · ScanNetV2 class-agnosticmAP: 33.4 -> 38.5Improved
merging threshold 0.5 instead of 0.9 (ablation)MV3DIS · ScanNetV2 class-agnosticmAP: 31.7 -> 38.5Improved
use MV3DIS instead of SAM2Objectzero-shot 3D instance segmentation · ScanNetV2 class-agnosticmAP: 34.0 -> 38.5Improved
zero-shot MV3DIS vs supervised Mask3DScanNet200 class-agnostic · ScanNet200 class-agnosticAP50: 51.2 -> 54.7Improved
zero-shot MV3DIS vs Mask3D trained on ScanNetV2ScanNet++ class-agnostic · ScanNet++ class-agnosticmAP: 9.9 -> 22.0Improved

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.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy