6-DOF GraspNet: Variational Grasp Generation for Object Manipulation

Mousavian, Arsalan; Eppner, Clemens; Fox, Dieter

doi:10.1109/iccv.2019.00299

Cited by 450 publications

(411 citation statements)

References 17 publications

Supporting

Mentioning

370

Contrasting

Unclassified

Order By: Relevance

“…Recent methods leverage data from 3D object reconstruction [54,22,43]. Grasp-Net [45] formulates the problem using a variational autoencoder which, given an input point cloud generates several grasp hypotheses, later refined by a second network. This Figure 2: Grasp affordance annotations.…”

Section: Related Workmentioning

confidence: 99%

“…Note that this is a significantly more complex problem than that of generating robotic grasps, as robot end-effectors have much less DoF than the human hand. For instance, very recently, Mousavian et al [45] introduced GraspNet to predict 6-DoF for object manipulation. Furthermore, the common practice in robotics is to use RGB-D cameras which, despite simplifying the process of modeling the geometry of the objects, do not have the versatility of standard RGB cameras.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

GanHand: Predicting Human Grasp Affordances in Multi-Object Scenes

Corona

Pumarola

Alenyà

et al. 2020

2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

146

104

View full text Add to dashboard Cite

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

GanHand: Predicting Human Grasp Affordances in Multi-Object Scenes

Corona

Pumarola

Alenyà

et al. 2020

2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

146

104

View full text Add to dashboard Cite

show abstract

“…poking). Most datadriven grasping algorithms at present can perform grasping detection in 6-degrees of freedom (6-DoF) with either closedloop feedback, which only utilizes top-down grasps in simple tabletop settings [74] [75], or open-loop feedback [76], [77].…”

Section: A Graspingmentioning

confidence: 99%

Review of Deep Reinforcement Learning-Based Object Grasping: Techniques, Open Challenges, and Recommendations

2020

View full text Add to dashboard Cite

The motivation behind our work is to review and analyze the most relevant studies on deep reinforcement learning-based object manipulation. Various studies are examined through a survey of existing literature and investigation of various aspects, namely, the intended applications, techniques applied, challenges faced by researchers and recommendations for minimizing obstacles. This review refers to all relevant articles on deep reinforcement learning-based object manipulation and solutions. The object grasping issue is a major manipulation challenge. Object grasping requires detection systems, methods and tools to facilitate efficient and fast agent training. Several studies have proposed that object grasping and its subtypes are the main elements in dealing with the environment and agent. Unlike other review articles, this review article provides different observations on deep reinforcement learning-based manipulation. The results of this comprehensive review of deep reinforcement learning in the manipulation field may be valuable for researchers and practitioners because they can expedite the establishment of important guidelines.

show abstract

“…Along this line, methods predict the success of a proposed grasp by training a traditional classifier (Jiang et al, 2011 ; Fischinger et al, 2015 ) or deep neural network (Saxena et al, 2008 ; Lenz et al, 2015 ; Redmon and Angelova, 2015 ; Pinto and Gupta, 2016 ; Kumra and Kanan, 2017 ; Wang et al, 2017 ). Alternatively, grasp simulation or analytical grasp metrics are computed for objects in model databases to generate training data (Johns et al, 2016 ; Mahler et al, 2016 , 2017 ; ten Pas et al, 2017 ; Cai et al, 2019 ; Liang et al, 2019 ; Mousavian et al, 2019 ). The task is then to learn a model that can predict the value of the grasp metric given a proposal and then select the grasp that is most likely to succeed.…”

Section: Related Workmentioning

confidence: 99%

DGCM-Net: Dense Geometrical Correspondence Matching Network for Incremental Experience-Based Robotic Grasping

2020

View full text Add to dashboard Cite

This article presents a method for grasping novel objects by learning from experience. Successful attempts are remembered and then used to guide future grasps such that more reliable grasping is achieved over time. To transfer the learned experience to unseen objects, we introduce the dense geometric correspondence matching network (DGCM-Net). This applies metric learning to encode objects with similar geometry nearby in feature space. Retrieving relevant experience for an unseen object is thus a nearest neighbor search with the encoded feature maps. DGCM-Net also reconstructs 3D-3D correspondences using the view-dependent normalized object coordinate space to transform grasp configurations from retrieved samples to unseen objects. In comparison to baseline methods, our approach achieves an equivalent grasp success rate. However, the baselines are significantly improved when fusing the knowledge from experience with their grasp proposal strategy. Offline experiments with a grasping dataset highlight the capability to transfer grasps to new instances as well as to improve success rate over time from increasing experience. Lastly, by learning task-relevant grasps, our approach can prioritize grasp configurations that enable the functional use of objects.

show abstract

6-DOF GraspNet: Variational Grasp Generation for Object Manipulation

Cited by 450 publications

References 17 publications

GanHand: Predicting Human Grasp Affordances in Multi-Object Scenes

GanHand: Predicting Human Grasp Affordances in Multi-Object Scenes

Review of Deep Reinforcement Learning-Based Object Grasping: Techniques, Open Challenges, and Recommendations

DGCM-Net: Dense Geometrical Correspondence Matching Network for Incremental Experience-Based Robotic Grasping

Contact Info

Product

Resources

About