AI Paper Atlas

Research map / 3D vision / 3D generation

VoxDet: Voxel Learning for Novel Instance Detection

Bowen Li, Jiashun Wang, Yaoyu Hu, Chen Wang, Sebastian Scherer

NeurIPS 2023 · 3D vision › 3D generation · approach: Learning 3D reconstruction from images · 1 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
Remove reconstruction pre-training (ablation)VoxDet · RoboToolsmAR: 18.7 -> 14.2Hurt
Replace voxel relation with 3D depth-wise convolution matching (ablation)VoxDet · RoboToolsmAR: 18.7 -> 15.1Hurt
Remove rotation measurement module R (ablation)VoxDet · RoboToolsmAR: 18.7 -> 15.6Hurt
Do not supervise the predicted rotation (ablation)VoxDet · RoboToolsmAR: 18.7 -> 18.2Hurt
Voxel matching instead of correlation-based 2D matchingVoxDet · LM-O/YCB-V averagemAR: 23.6 -> 30.4Improved
Voxel matching instead of DINO cosine similarityVoxDet · LM-O/YCB-V averagemAR: 24.6 -> 30.4Improved
Voxel matching instead of correlationVoxDet · RoboTools (sim-to-real)mAR: 14.4 -> 18.7Improved
Voxel matching instead of correlation matchingVoxDet · per-image inferencetotal time (s): 0.182 -> 0.154Improved
Voxel matching instead of DINO featuresVoxDet · per-image inferencetotal time (s): 0.357 -> 0.154Improved
Reduce templates from 10 to 2OLN_DINO 2D baseline vs VoxDet · YCB-Vaccuracy drop: 87%Hurt
Reduce Top-K from 100 to 10VoxDet vs OS2D · Top-K analysisAR: 5~10% vs 38%Mixed

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