AI Paper Atlas

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

ICCV 2025 · 3D vision › 3D generation · approach: Learning 3D reconstruction from images · 16 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 DINOv2 2D features to MUSt3R 3D features (ablation)PanSt3R · ScanNet++ val (224, +LUDVIG)PQ: 46.4 -> 50.4Improved
add MUSt3R 3D features to DINOv2 2D features (ablation)PanSt3R · ScanNet++ val (224, +LUDVIG)PQ: 35.7 -> 50.4Improved
add QUBO mask merging (ablation)PanSt3R + LUDVIG (no reg) · HypersimPQ: 58.1 -> 66.7Improved
add QUBO mask merging (ablation)PanSt3R + LUDVIG (no reg) · ScanNetPQ: 60.2 -> 67.3Improved
add QUBO mask merging (ablation)PanSt3R + LUDVIG (no reg) · ReplicaPQ: 60.8 -> 60.7No clear effect
add QUBO mask merging to direct prediction (ablation)PanSt3R (rendered views) · HypersimPQ: 51.6 -> 56.5Improved
add QUBO mask merging to direct prediction (ablation)PanSt3R (rendered views) · ScanNetPQ: 59.5 -> 65.7Improved
add panoptic regularization to 3DGS (QUBO off) (ablation)PanSt3R + LUDVIG · ScanNet++ valPQ: 50.8 -> 55.2Improved
add panoptic regularization to 3DGS (QUBO on) (ablation)PanSt3R + LUDVIG · ScanNet++ valPQ: 52.0 -> 54.7Hurt
add panoptic regularization to 3DGS (QUBO on) (ablation)PanSt3R + LUDVIG · HypersimPQ: 66.7 -> 66.3Hurt
use 50 keyframes instead of 10 (ablation)PanSt3R · ScanNet++ seen setPQ: 49.3 -> 58.2Improved
use 100 keyframes instead of 50 (ablation)PanSt3R · ScanNet++ seen setPQ: 58.2 -> 58.2No 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.

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.

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