DexYCB: A Benchmark for Capturing Hand Grasping of Objects

被引：115

作者：

Chao, Yu-Wei ^{[1
]}

Yang, Wei ^{[1
]}

Xiang, Yu ^{[1
]}

Molchanov, Pavlo ^{[1
]}

Handa, Ankur ^{[1
]}

Tremblay, Jonathan ^{[1
]}

Narang, Yashraj S. ^{[1
]}

Van Wyk, Karl ^{[1
]}

Iqbal, Umar ^{[1
]}

Birchfield, Stan ^{[1
]}

Kautz, Jan ^{[1
]}

Fox, Dieter ^{[1
,2
]}

机构：

[1] NVIDIA, Santa Clara, CA 95050 USA

[2] Univ Washington, Seattle, WA 98195 USA

来源：

2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021 | 2021年

关键词：

D O I：

10.1109/CVPR46437.2021.00893

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

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.(1)

引用

页码：9040 / 9049

页数：10

共 50 条

[1] [Anonymous], 2017, SIGGRAPH ASIA, DOI DOI 10.1145/3130800.3130883
[2] [Anonymous], 2019, CVPR, DOI DOI 10.1109/CVPR.2019.00469
[3] [Anonymous], 2019, CVPR, DOI DOI 10.1109/CVPR.2019.01110
[4] [Anonymous], 2020, CVPR, DOI DOI 10.1109/CVPR42600.2020.00664
[5] Armagan Anil, 2020, ECCV
[6] Brahmbhatt S., 2020, P EUR C COMP VIS, P361
[7] CORONA E, 2020, CVPR, P5030, DOI DOI 10.1109/CVPR42600.2020.00508
[8] Dandan Shan, 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Proceedings, P9866, DOI 10.1109/CVPR42600.2020.00989
[9] Deng XK, 2019, ROBOTICS: SCIENCE AND SYSTEMS XV
[10] Eppner C., 2019, ISRR

← 1 2 3 4 5 →