Region-aware Grasp Framework with Normalized Grasp Space for Efficient 6-DoF Grasping

Siang Chen, Pengwei Xie, Wei Tang, Dingchang Hu, Yixiang Dai, Guijin Wang
Proceedings of The 8th Conference on Robot Learning, PMLR 270:1834-1850, 2025.

Abstract

A series of region-based methods succeed in extracting regional features and enhancing grasp detection quality. However, faced with a cluttered scene with potential collision, the definition of the grasp-relevant region stays inconsistent. In this paper, we propose Normalized Grasp Space (NGS) from a novel region-aware viewpoint, unifying the grasp representation within a normalized regional space and benefiting the generalizability of methods. Leveraging the NGS, we find that CNNs are underestimated for 3D feature extraction and 6-DoF grasp detection in clutter scenes and build a highly efficient Region-aware Normalized Grasp Network (RNGNet). Experiments on the public benchmark show that our method achieves significant >20 % performance gains while attaining a real-time inference speed of approximately 50 FPS. Real-world cluttered scene clearance experiments underscore the effectiveness of our method. Further, human-to-robot handover and dynamic object grasping experiments demonstrate the potential of our proposed method for closed-loop grasping in dynamic scenarios.

Cite this Paper

BibTeX
@InProceedings{pmlr-v270-chen25c, title = {Region-aware Grasp Framework with Normalized Grasp Space for Efficient 6-DoF Grasping}, author = {Chen, Siang and Xie, Pengwei and Tang, Wei and Hu, Dingchang and Dai, Yixiang and Wang, Guijin}, booktitle = {Proceedings of The 8th Conference on Robot Learning}, pages = {1834--1850}, year = {2025}, editor = {Agrawal, Pulkit and Kroemer, Oliver and Burgard, Wolfram}, volume = {270}, series = {Proceedings of Machine Learning Research}, month = {06--09 Nov}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v270/main/assets/chen25c/chen25c.pdf}, url = {https://proceedings.mlr.press/v270/chen25c.html}, abstract = {A series of region-based methods succeed in extracting regional features and enhancing grasp detection quality. However, faced with a cluttered scene with potential collision, the definition of the grasp-relevant region stays inconsistent. In this paper, we propose Normalized Grasp Space (NGS) from a novel region-aware viewpoint, unifying the grasp representation within a normalized regional space and benefiting the generalizability of methods. Leveraging the NGS, we find that CNNs are underestimated for 3D feature extraction and 6-DoF grasp detection in clutter scenes and build a highly efficient Region-aware Normalized Grasp Network (RNGNet). Experiments on the public benchmark show that our method achieves significant >20 % performance gains while attaining a real-time inference speed of approximately 50 FPS. Real-world cluttered scene clearance experiments underscore the effectiveness of our method. Further, human-to-robot handover and dynamic object grasping experiments demonstrate the potential of our proposed method for closed-loop grasping in dynamic scenarios.} }
Endnote
%0 Conference Paper %T Region-aware Grasp Framework with Normalized Grasp Space for Efficient 6-DoF Grasping %A Siang Chen %A Pengwei Xie %A Wei Tang %A Dingchang Hu %A Yixiang Dai %A Guijin Wang %B Proceedings of The 8th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2025 %E Pulkit Agrawal %E Oliver Kroemer %E Wolfram Burgard %F pmlr-v270-chen25c %I PMLR %P 1834--1850 %U https://proceedings.mlr.press/v270/chen25c.html %V 270 %X A series of region-based methods succeed in extracting regional features and enhancing grasp detection quality. However, faced with a cluttered scene with potential collision, the definition of the grasp-relevant region stays inconsistent. In this paper, we propose Normalized Grasp Space (NGS) from a novel region-aware viewpoint, unifying the grasp representation within a normalized regional space and benefiting the generalizability of methods. Leveraging the NGS, we find that CNNs are underestimated for 3D feature extraction and 6-DoF grasp detection in clutter scenes and build a highly efficient Region-aware Normalized Grasp Network (RNGNet). Experiments on the public benchmark show that our method achieves significant >20 % performance gains while attaining a real-time inference speed of approximately 50 FPS. Real-world cluttered scene clearance experiments underscore the effectiveness of our method. Further, human-to-robot handover and dynamic object grasping experiments demonstrate the potential of our proposed method for closed-loop grasping in dynamic scenarios.
APA
Chen, S., Xie, P., Tang, W., Hu, D., Dai, Y. & Wang, G.. (2025). Region-aware Grasp Framework with Normalized Grasp Space for Efficient 6-DoF Grasping. Proceedings of The 8th Conference on Robot Learning, in Proceedings of Machine Learning Research 270:1834-1850 Available from https://proceedings.mlr.press/v270/chen25c.html.

Related Material