End-to-End Dexterous Grasp Learning from Single-View Point Clouds via a Multi-Object Scene Dataset

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Geng, Tao, Yang, Dapeng, Liu, Ziwei, Zhang, Le, Qi, Le, Li, WangYang, Ren, Yi, Luo, Shan, Ni, Fenglei
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911519936086016
author Geng, Tao
Yang, Dapeng
Liu, Ziwei
Zhang, Le
Qi, Le
Li, WangYang
Ren, Yi
Luo, Shan
Ni, Fenglei
author_facet Geng, Tao
Yang, Dapeng
Liu, Ziwei
Zhang, Le
Qi, Le
Li, WangYang
Ren, Yi
Luo, Shan
Ni, Fenglei
contents Dexterous grasping in multi-object scene constitutes a fundamental challenge in robotic manipulation. Current mainstream grasping datasets predominantly focus on single-object scenarios and predefined grasp configurations, often neglecting environmental interference and the modeling of dexterous pre-grasp gesture, thereby limiting their generalizability in real-world applications. To address this, we propose DGS-Net, an end-to-end grasp prediction network capable of learning dense grasp configurations from single-view point clouds in multi-object scene. Furthermore, we propose a two-stage grasp data generation strategy that progresses from dense single-object grasp synthesis to dense scene-level grasp generation. Our dataset comprises 307 objects, 240 multi-object scenes, and over 350k validated grasps. By explicitly modeling grasp offsets and pre-grasp configurations, the dataset provides more robust and accurate supervision for dexterous grasp learning. Experimental results show that DGS-Net achieves grasp success rates of 88.63\% in simulation and 78.98\% on a real robotic platform, while exhibiting lower penetration with a mean penetration depth of 0.375 mm and penetration volume of 559.45 mm^3, outperforming existing methods and demonstrating strong effectiveness and generalization capability. Our dataset is available at https://github.com/4taotao8/DGS-Net.
format Preprint
id arxiv_https___arxiv_org_abs_2603_15410
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle End-to-End Dexterous Grasp Learning from Single-View Point Clouds via a Multi-Object Scene Dataset
Geng, Tao
Yang, Dapeng
Liu, Ziwei
Zhang, Le
Qi, Le
Li, WangYang
Ren, Yi
Luo, Shan
Ni, Fenglei
Robotics
Dexterous grasping in multi-object scene constitutes a fundamental challenge in robotic manipulation. Current mainstream grasping datasets predominantly focus on single-object scenarios and predefined grasp configurations, often neglecting environmental interference and the modeling of dexterous pre-grasp gesture, thereby limiting their generalizability in real-world applications. To address this, we propose DGS-Net, an end-to-end grasp prediction network capable of learning dense grasp configurations from single-view point clouds in multi-object scene. Furthermore, we propose a two-stage grasp data generation strategy that progresses from dense single-object grasp synthesis to dense scene-level grasp generation. Our dataset comprises 307 objects, 240 multi-object scenes, and over 350k validated grasps. By explicitly modeling grasp offsets and pre-grasp configurations, the dataset provides more robust and accurate supervision for dexterous grasp learning. Experimental results show that DGS-Net achieves grasp success rates of 88.63\% in simulation and 78.98\% on a real robotic platform, while exhibiting lower penetration with a mean penetration depth of 0.375 mm and penetration volume of 559.45 mm^3, outperforming existing methods and demonstrating strong effectiveness and generalization capability. Our dataset is available at https://github.com/4taotao8/DGS-Net.
title End-to-End Dexterous Grasp Learning from Single-View Point Clouds via a Multi-Object Scene Dataset
topic Robotics
url https://arxiv.org/abs/2603.15410