Great success has been achieved in the 6-DoF grasp learning from the point cloud input, yet the computational cost due to the point set orderlessness remains a concern. Alternatively, we explore the grasp generation from the RGB-D input in this paper. The proposed solution, Keypoint-GraspNet, detects the projection of the gripper keypoints in the image space and then recover the SE(3) poses with a PnP algorithm. A synthetic dataset based on the primitive shape and the grasp family is constructed to examine our idea. Metric-based evaluation reveals that our method outperforms the baselines in terms of the grasp proposal accuracy, diversity, and the time cost. Finally, robot experiments show high success rate, demonstrating the potential of the idea in the real-world applications.

翻译：从点云输入进行6-DoF抓取学习已取得巨大成功，但点集无序性带来的计算开销仍是一个问题。本文另辟蹊径，探索从RGB-D输入直接生成抓取姿态。所提方案Keypoint-GraspNet首先检测图像空间中夹持器关键点的投影，随后通过PnP算法恢复SE(3)姿态。我们基于基本几何形状与抓取族构建合成数据集以验证该思路。基于指标的评估表明，本方法在抓取提案精度、多样性及时间开销方面均优于基线方法。最后，机器人实验获得了较高成功率，验证了该思路在现实应用中的潜力。