AI Paper Atlas

Research map / 3D vision / 3D generation

RGB-Pointmap Pretraining for Unified 3D Scene Understanding

Ye Mao, Weixun Luo, Ranran Huang, Junpeng Jing, Krystian Mikolajczyk

ECCV 2026 · 3D vision › 3D generation · approach: Semantic scene completion · 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 pointmap input to image-only encoder (ablation)UniScene3D · ScanRefer viewpoint groundingR@1: 28.8 -> 38.6Improved
add image input to pointmap-only encoder (ablation)UniScene3D · ScanRefer viewpoint groundingR@1: 37.6 -> 38.6Improved
add image input to pointmap-only encoder (ablation)UniScene3D · ScanNet scene type classification (0-shot)accuracy: 67.2 -> 70.7Improved
add image input to pointmap-only encoder (ablation)UniScene3D · ScanNet scene type classification (10-shot)accuracy: 83.6 -> 83.7No clear effect
add grounded view alignment loss (ablation)UniScene3D · ScanRefer viewpoint groundingR@1: 18.5 -> 38.6Improved
add grounded view alignment loss (ablation)UniScene3D · Sr3D viewpoint groundingR@1: 11.5 -> 23.6Improved
add cross-view geometric alignment loss (ablation)UniScene3D · ScanRefer viewpoint groundingR@1: 38.2 -> 38.6Improved
add cross-view geometric alignment loss (ablation)UniScene3D · Nr3D viewpoint groundingR@1: 23.2 -> 25.2Improved
add cross-view geometric alignment loss (ablation)UniScene3D · ScanNet scene type classification (10-shot)accuracy: 81.3 -> 83.7Improved
add grounded view alignment loss (ablation)UniScene3D · ScanRefer scene retrieval (n=10)R@5: 63.0 -> 62.9Hurt
use summation early fusion instead of concat + projection (ablation)UniScene3D · ScanRefer viewpoint groundingR@1: 36.5 -> 38.6Improved
initialize pointmap patch embedding from image patch embedding (ablation)UniScene3D · ScanRefer viewpoint groundingR@1: 38.0 -> 38.6Improved

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