AI Paper Atlas

Research map / 3D vision / Human pose estimation

CRAVES: Controlling Robotic Arm With a Vision-Based Economic System

Yiming Zuo, Weichao Qiu, Lingxi Xie, Fangwei Zhong, Yizhou Wang, Alan L. Yuille

CVPR 2019 · 3D vision › Human pose estimation · approach: Depth-based hand pose regression · 55 citations (OpenAlex)

Check 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 geometric constraints to refine keypoints and fine-tune on unlabeled real imagesSynthetic-pretrained keypoint model · YouTube datasetPCK@0.2 (%): 80.05 -> 87.01Improved
Use geometric constraints to refine keypoints and fine-tune on unlabeled real imagesSynthetic-pretrained keypoint model · Lab datasetPCK@0.2 (%): 95.66 -> 99.55Improved
Use geometric constraints to refine keypoints and fine-tune on unlabeled real imagesSynthetic-pretrained pose estimator · Lab datasetAverage joint angular error (degrees): 7.13 -> 4.81Improved
Add background augmentation during fine-tuning (ablation)Real-image keypoint model · Lab datasetPCK@0.2 (%): 94.52 -> 98.72Improved
Add background augmentation during fine-tuning (ablation)Real-image keypoint model · YouTube datasetPCK@0.2 (%): 80.24 -> 84.04Improved
Add joint keypoint detection and 3D pose estimation (ablation)Real-image keypoint model · YouTube datasetPCK@0.2 (%): 80.24 -> 86.52Improved
Add joint keypoint detection and 3D pose estimation (ablation)Real-image keypoint model · Lab datasetPCK@0.2 (%): 94.52 -> 97.31Improved
Combine background augmentation with joint keypoint detection and 3D pose estimation (ablation)Real-image keypoint model · YouTube datasetPCK@0.2 (%): 80.24 -> 87.01Improved
Increase synthetic training images from 2,500 to 5,000 (ablation)Synthetic-pretrained keypoint model · YouTube datasetPCK@0.2 (%): 75.24 -> 80.05Improved
Increase synthetic training images from 5,000 to 20,000 (ablation)Synthetic-pretrained keypoint model · YouTube datasetPCK@0.2 (%): 80.05 -> 81.11Improved
Use the vision-based controller rather than human control from camera inputReaching task · Reaching taskSuccess rate: 66.7% -> 59.3%Hurt
Use the vision-based controller rather than human control from camera inputReaching task · Reaching taskAverage time (s): 38.8 -> 21.2Improved

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