6-DoF Task Grasp (6DTG)
收藏资源简介:
6-DoF Task Grasp (6DTG) 数据集是由中山大学计算机科学与工程学院构建的大型6-DoF任务导向抓取数据集。该数据集包含4391个杂乱场景,超过200万个6-DoF抓取姿态,每个抓取姿态都标注了特定任务。数据集的特点包括:细粒度的抓取任务标注、6-DoF抓取姿态和杂乱场景、密集标注的抓取姿态。该数据集旨在解决任务导向的6-DoF抓取姿态检测问题,为机器人抓取研究提供更实际和通用的数据支持。
The 6-DoF Task Grasp (6DTG) dataset is a large-scale 6-DoF task-oriented grasping dataset constructed by the School of Computer Science and Engineering, Sun Yat-sen University. It contains 4391 cluttered scenes and over 2 million 6-DoF grasping poses, with each grasping pose annotated with a specific task. The dataset is characterized by fine-grained grasping task annotations, 6-DoF grasping poses in cluttered scenes, and densely annotated grasping poses. This dataset aims to address the task-oriented 6-DoF grasping pose detection problem, providing more practical and general data support for robotic grasping research.

- 1Task-Oriented 6-DoF Grasp Pose Detection in Clutters中山大学计算机科学与工程学院 · 2025年



