AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Find Any Part in 3D

Ziqi Ma, Yisong Yue, Georgia Gkioxari

ICCV 2025 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary 3D and world perception · 13 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
evaluate zero-shot Find3D versus best baseline PointCLIPV2 on unseen categoriesPointCLIPV2 · Objaverse-General UnseenmIoU: 10.27 -> 26.21 (260% improvement in mIoU)Improved
change prompt from top-k to {part} of a {object} for PointCLIPV2PointCLIPV2 · ShapeNet-PartmIoU: 48.47 -> 17.42 (64% decrease)Hurt
random rotation of test objects for PointCLIPV2PointCLIPV2 · ShapeNet-PartmIoU: 46% dropHurt
random rotation of test objects for Find3DFind3D · ShapeNet-PartmIoU: 28.39 -> 29.64Improved
domain shift to ShapeNetPart-V2 for PointCLIPV2PointCLIPV2 · ShapeNetPart-V2mIoU: 48.47 to 21.18 (56% drop)Hurt
domain shift to ShapeNetPart-V2 for Find3DFind3D · ShapeNetPart-V2mIoU: 20% increaseImproved
domain shift to ShapeNetPart-V2 for closed-world PointNeXtPointNeXt · ShapeNetPart-V2mIoU: 80.44 -> 28.70 (64% drop)Hurt
vary query prompt for PartSLIP++ on PartNet-EPartSLIP++ · PartNet-EmIoU: 32.71 -> 5.12 (84% drop)Hurt
scale training object categories from 16 to 761 with the data engine (ablation)Find3D · Objaverse-General UnseenmIoUImproved
use top-k prompt search instead of direct promptsPointCLIPV2 vs Find3D · ShapeNet-Part canonicalmIoU: PointCLIPV2 48.666 vs Find3D 43.613Mixed
Find3D zero-shot vs PartSLIP++ trained on PartNet-E with part promptPartSLIP++ · PartNet-EmIoU: 32.71 vs 16.38Hurt
Find3D feed-forward vs PartSLIP++ 2D-3D aggregationPartSLIP++ · PartNet-E subsetinference time: 174.3s -> 0.9sImproved

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