AI Paper Atlas

Research map / 3D vision / 3D generation

Rel3D: A Minimally Contrastive Benchmark for Grounding Spatial Relations in 3D

Ankit Goyal, Kaiyu Yang, Dawei Yang, Jia Deng

NeurIPS 2020 · 3D vision › 3D generation · approach: Open-vocabulary 3D scene understanding · 4 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
Train on minimally contrastive pairs rather than non-contrastive examples (ablation)DRNet, VTransE, and 3D-feature models · Rel3Dsample efficiency: only about $1/4 training dataImproved
Use a simple 2D bounding-box baseline instead of sophisticated visual relationship modelsRel3D spatial relation recognition · Rel3Daverage class accuracyImproved
Use language-only input instead of visual or geometric inputRel3D spatial relation recognition · Rel3Daccuracy: 50%No clear effect
Use aligned 3D features instead of raw 3D features (ablation)MLP spatial relation classifier · Rel3Daccuracy: 0.803 vs 0.682 (behind); 0.698 vs 0.708 (passing through)Mixed
Use 3D position, pose, and scale featuresMLP spatial relation classifier · Rel3Dspatial relation recognitionImproved
Rely on 3D features alone as a complete solutionMLP spatial relation classifier · Rel3Dspatial relation recognitionHurt

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