AI Paper Atlas

Research map / 3D vision / Human pose estimation

Breaking the 3D Dataset Bottleneck: Fast Scalable Generation of Aligned 3D Assets from Scratch for Category 6D Pose Estimation and Robotic Grasping

Duret Guillaume, Danylo Mazurak, Florence Zara, Jan Peters, Liming Chen

CVPR 2026 · 3D vision › Human pose estimation · approach: Instance-level 6D object pose · 0 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 depth conditioning (ControlNet) to text-to-image generation (ablation)text-to-image-to-3D pipeline · NOCS categoriespose consistency (%): 57 -> 97Improved
add depth conditioning for laptops (ablation)image generation · NOCS Laptoppose consistency (%): 20 -> 90Improved
add shadows to mixed-reality training data (generated meshes) (ablation)DualPoseNet · NOCS REAL275 zero-shotaverage pose score: 30.83 -> 34.75Improved
add shadows to mixed-reality training data (original NOCS meshes) (ablation)DualPoseNet · NOCS REAL275 zero-shotaverage pose score: 29.07 -> 33.24Improved
replace original NOCS meshes with generated meshes (with shadows) (ablation)DualPoseNet · NOCS REAL275 zero-shotaverage pose score: 33.24 -> 34.75Improved
replace original NOCS meshes with generated meshes (no shadows) (ablation)DualPoseNet · NOCS REAL275 zero-shotaverage pose score: 29.07 -> 30.83Improved
replace original NOCS meshes with generated meshes (ablation)DualPoseNet · synthetic validationaverage pose score: 15.66 -> 23.91Improved
use mixed-reality rendering instead of fully synthetic Replica scenes (ablation)DualPoseNet · NOCS REAL275 zero-shotaverage pose score: 33.10 -> 34.75Improved
train CenterGrasp on generated custom data instead of pretrained CenterGraspCenterGrasp · SAPIEN grasping (textured eval)grasp success rate: 0.7896 -> 0.8679Improved
train on generated custom dataCenterGrasp · SAPIEN shape completion (textured eval)IoU: 0.314 -> 0.475Improved
train with native textures instead of randomized textures, tested on randomized textures (ablation)Custom-CG · SAPIEN grasping (randomized eval)grasp success rate: 0.8784 -> 0.8264Hurt
use specialized category-fine-tuned model instead of baseline CenterGraspCenterGrasp · real robot, Bowlgrasp success (%): 14% -> 57%Improved

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