AI Paper Atlas

Research map / 3D vision / Face and shape

3DCoMPaT ++ : An Improved Large-Scale 3D Vision Dataset for Compositional Recognition

Habib Slim, Xiang Li, Yuchen Li, Mahmoud Ahmed, Mohamed Ayman, Ujjwal Upadhyay, Ahmed Abdelreheem, Arpit Prajapati, Suhail Pothigara, Peter Wonka, Mohamed Elhoseiny

TPAMI 2025 · 3D vision › Face and shape · approach: Single-view 3D reconstruction · 15 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 shape-category prior in training and inferenceCurveNet · fine-grained 3D part segmentationshape-agnostic mIOU: 53.09 -> 59.61Improved
add shape-category prior in training and inferencePCT · fine-grained 3D part segmentationshape-agnostic mIOU: 49.09 -> 56.14Improved
add shape-category prior in training and inferenceCurveNet · coarse-grained 3D part segmentationshape-agnostic mIOU: 76.32 -> 83.85Improved
use coarse-grained instead of fine-grained part labelsCurveNet · 3D part segmentation (no shape prior)shape-agnostic mIOU: 53.09 -> 76.32Improved
provide ground-truth shape category (shape-informed mIOU) instead of shape-agnostic3D part segmentation models · fine-grained 3D part segmentationmIOU: 53.09 -> 81.37Improved
use ResNet-50 instead of ResNet-182D shape classification · 3DCoMPaT++ renderstop-1 accuracy: 76.27 -> 90.20Improved
use 2D/3D bidirectional-projection fusion (BPNet) instead of late fusion of separate 2D and 3D modelsGCR baseline · fine-grained GCRGrounded-value-all: 3.84 -> 15.41Improved
use RGB 6D point input with PointNet++ instead of BPNetGCR · coarse-grained GCRGrounded-value-all: 50.44 -> 58.99Improved
use RGB PointNet++ instead of BPNetGCR · fine-grained GCRValue-all: 27.74 -> 25.36Hurt
increase number of style compositions per shape from 1 to 5 and 20 (ablation)BPNet · GCR compositional metricsValue-all and grounded metricsImproved
increase number of style compositions from 20 to 50 (ablation)BPNet · GCR compositional metricsValue-all and grounded metricsNo clear effect

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