AI Paper Atlas

Research map / 3D vision / 3D generation

Hierarchical 3D Scene Graphs Construction Outdoors

Jon Nyffeler, Federico Tombari, Daniel Barath

ICCV 2025 · 3D vision › 3D generation · approach: Open-vocabulary 3D scene understanding · 2 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 structure (S) modality to point cloud (P) training (ablation)SGAligner on our scene graphs · LaMAR node matching, trained on LINmean reciprocal rank: 0.963 -> 0.939Hurt
add structure (S) modality to point cloud (P) training (ablation)SGAligner on our scene graphs · LaMAR node matching, trained on CABmean reciprocal rank: 0.944 -> 0.865Hurt
add structure (S) modality to point cloud (P) training (ablation)SGAligner on our scene graphs · LaMAR node matching, trained on HGEmean reciprocal rank: 0.942 -> 0.892Hurt
add hyperedge (H) modality to point cloud (P) training (ablation)SGAligner on our scene graphs · LaMAR node matching, trained on LINmean reciprocal rank: 0.963 -> 0.967Improved
add hyperedge (H) modality to point cloud (P) training (ablation)SGAligner on our scene graphs · LaMAR node matching, trained on CABmean reciprocal rank: 0.944 -> 0.938Hurt
add hyperedge (H) modality to point cloud (P) training (ablation)SGAligner on our scene graphs · LaMAR node matching, trained on HGEmean reciprocal rank: 0.942 -> 0.944Improved
use hyperedge plus attribute (H + A) modalities instead of point cloud only (ablation)SGAligner on our scene graphs · LaMAR node matching, trained on LINmean reciprocal rank: 0.963 -> 0.996Improved
use hyperedge plus attribute (H + A) modalities instead of point cloud only (ablation)SGAligner on our scene graphs · LaMAR node matching, trained on HGEmean reciprocal rank: 0.942 -> 0.971Improved
use hyperedge plus attribute (H + A) modalities instead of point cloud only (ablation)SGAligner on our scene graphs · LaMAR node matching, trained on CABmean reciprocal rank: 0.944 -> 0.941No clear effect
replace direct FPFH registration with SGAligner object matching on our scene graphs (FPFH)FPFH registration · LaMAR LIN point cloud registrationRRE (degrees): 32.049 -> 0.279Improved
replace direct FPFH registration with SGAligner object matching on our scene graphs (FPFH)FPFH registration · LaMAR CAB point cloud registrationRTE (meters): 14.453 -> 1.552Improved
replace direct FPFH registration with SGAligner object matching on our scene graphs (FPFH)FPFH registration · LaMAR HGE point cloud registrationruntime (s): 2.644 -> 5.981Hurt

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