AI Paper Atlas

Research map / 3D vision / Human pose estimation

Markerless Camera-to-Robot Pose Estimation via Self-Supervised Sim-to-Real Transfer

Jingpei Lu, Florian Richter, Michael C. Yip

CVPR 2023 · 3D vision › Human pose estimation · approach: Learned camera pose estimation · 29 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
Use CtRNet instead of DREAM or RoboPoseCamera-to-robot pose estimation · DREAM-real AllAUC ADD: 85.962 vs 56.988 for DREAM-Q; 85.962 vs 80.094 for RoboPoseImproved
Use CtRNet instead of Diff. RenderingCamera-to-robot pose estimation · BaxterPCK AUC: 93.94 vs 78.60Improved
Use CtRNet instead of Diff. RenderingCamera-to-robot pose estimation · BaxterADD AUC: 83.93 vs 81.15Improved
Use CtRNet instead of DREAMBaxter visual servoing · Baxter visual servoingTrans. Err. (m): 0.235 ± 0.313 -> 0.002 ± 0.001Improved
Increase synthetic pretraining samples from 500 to 1000 (ablation)CtRNet · BaxterAUC ADD: 47.62 -> 76.65Improved
Increase synthetic pretraining samples from 1000 to 2000 (ablation)CtRNet · BaxterAUC ADD: 76.65 -> 82.98Improved
Increase synthetic pretraining samples beyond 2000 (ablation)CtRNet · BaxterAUC ADD: 82.98 -> 84.12 -> 83.93Mixed
Add segmentation loss L_seg (ablation)CtRNet with 500 pretraining samples · BaxterAUC ADD: 65.26 -> 47.62Hurt
Add segmentation loss L_seg (ablation)CtRNet with 1000 pretraining samples · BaxterAUC ADD: 69.71 -> 76.65Improved
Add segmentation loss L_seg (ablation)CtRNet with 2000 pretraining samples · BaxterAUC ADD: 76.20 -> 82.98Improved
Add segmentation loss L_seg (ablation)CtRNet with 4000 pretraining samples · BaxterAUC ADD: 78.71 -> 84.12Improved
Add segmentation loss L_seg (ablation)CtRNet with 8000 pretraining samples · BaxterAUC ADD: 79.07 -> 83.93Improved

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