1. [Publications](/publications)
2. DexYCB: A Benchmark for Capturing Hand Grasping of Objects
 
 # DexYCB: A Benchmark for Capturing Hand Grasping of Objects

  ![](/sites/default/files/publications/dex-ycb-render_1280x0720.gif) 

 We introduce DexYCB, a new dataset for capturing hand grasping of objects. We first compare DexYCB with a related one through cross-dataset evaluation. We then present a thorough benchmark of state-of-the-art approaches on three relevant tasks: 2D object and keypoint detection, 6D object pose estimation, and 3D hand pose estimation. Finally, we evaluate a new robotics-relevant task: generating safe robot grasps in human-to-robot object handover.



 ## Authors



[Yu-Wei Chao](/person/yu-wei-chao)

[Wei Yang](/person/wei-yang)

Yu Xiang (NVIDIA)

[Pavlo Molchanov](/person/pavlo-molchanov)

Ankur Handa (NVIDIA)

[Jonathan Tremblay](/person/jonathan-tremblay)

[Yashraj Narang](/person/yashraj-narang)

Karl Van Wyk (NVIDIA)

[Umar Iqbal](/person/umar-iqbal)

[Stan Birchfield](/person/stan-birchfield)

[Jan Kautz](/person/jan-kautz)

Dieter Fox (NVIDIA)

 

 

 ## Publication Date



Monday, June 7, 2021

 

 ## Published in



[IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2021](http://cvpr2021.thecvf.com)

 

 ## Research Area



[Computer Vision](/research-area/computer-vision)

[Robotics](/research-area/robotics)

 

 

 ## External Links



[Paper, Data, Code, Video](https://dex-ycb.github.io)