Knowledge Commons of Institute of Automation,CAS
HGCNet: Deep Anthropomorphic Hand Grasping in Clutter | |
Li YM(李一鸣) | |
2022-05 | |
会议名称 | IEEE International Conference on Robotics and Automation |
会议日期 | 2022-5 |
会议地点 | 线上+线下(美国费城) |
出版者 | IEEE |
摘要 | Grasping in cluttered environments is one of the most fundamental skills in robotic manipulation. Most of the current works focus on estimating grasp poses for parallel-jaw or suction-cup end effectors. However, the study for dexterous anthropomorphic hand grasping in clutter remains a great challenge. In this paper, we propose HGC-Net, a single-shot network that learns to predict dense hand grasp configurations in clutter from single-view point cloud input. Our end-to-end neural network can predict hand grasp proposals efficiently and effectively. To enhance generalization, we built a largescale synthetic grasping dataset with 179 household objects, 5K cluttered scenes and over 10M hand annotations. Experiments in simulation show that our model can predict dense and robust hand grasps and clear over 78% of unseen objects in clutter without any post-processing and outperform baseline methods by a large margin. Experiments on the real robot platform also demonstrate that the model trained on synthetic data performs well in natural environments. Code is available at https://github.com/yimingli1998/hgc net. |
收录类别 | EI |
语种 | 英语 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/48751 |
专题 | 智能机器人系统研究 |
作者单位 | 1.中国科学院自动化研究所 2.中国科学院大学 |
推荐引用方式 GB/T 7714 | Li YM. HGCNet: Deep Anthropomorphic Hand Grasping in Clutter[C]:IEEE,2022. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
HGC_Net (1).pdf(5779KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
个性服务 |
推荐该条目 |
保存到收藏夹 |
查看访问统计 |
导出为Endnote文件 |
谷歌学术 |
谷歌学术中相似的文章 |
[Li YM(李一鸣)]的文章 |
百度学术 |
百度学术中相似的文章 |
[Li YM(李一鸣)]的文章 |
必应学术 |
必应学术中相似的文章 |
[Li YM(李一鸣)]的文章 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论