Filters








22,182 Hits in 7.1 sec

Generalized Feedback Loop for Joint Hand-Object Pose Estimation [article]

Markus Oberweger, Paul Wohlhart, Vincent Lepetit
2019 arXiv   pre-print
Our approach performs en-par with state-of-the-art methods for 3D hand pose estimation, and outperforms state-of-the-art methods for joint hand-object pose estimation when using depth images only.  ...  The components of this feedback loop are also Deep Networks, optimized using training data. This approach can be generalized to a hand interacting with an object.  ...  ACKNOWLEDGMENTS The authors would like to thank Zeyuan Chen for his help with the synthetic hand dataset. Prof. Vincent Lepetit is a Senior Member of the Institut Universitaire de France.  ... 
arXiv:1903.10883v1 fatcat:vkxyqrpitbf3hh4dvnd4mhdkwm

kPAM 2.0: Feedback Control for Category-Level Robotic Manipulation [article]

Wei Gao, Russ Tedrake
2021 arXiv   pre-print
Previous approaches typically build a feedback loop on top of a real-time 6-DOF pose estimator.  ...  With the proposed object and action representation, our framework is also agnostic to the robot grasp pose and initial object configuration, making it flexible for integration and deployment.  ...  They build a feedback loop on top of a real-time pose estimator.  ... 
arXiv:2102.06279v1 fatcat:alxl4vojk5annoyjvp6cnziyoa

Online Body Schema Adaptation Based on Internal Mental Simulation and Multisensory Feedback

Pedro Vicente, Lorenzo Jamone, Alexandre Bernardino
2016 Frontiers in Robotics and AI  
Predictions obtained through a mental simulation of the body are combined with the real sensory feedback to achieve two objectives simultaneously: body schema adaptation and markerless 6D hand pose estimation  ...  The updated body schema will improve the estimates of the 6D hand pose, which is then used in a closed-loop control scheme (i.e., visual servoing), enabling precise reaching.  ...  estimation of the end-effector pose and (ii) during the closed-loop control (described in Section "Closed-Loop"), the end-effector pose feedback is exploited to accurately reach for the desired pose.  ... 
doi:10.3389/frobt.2016.00007 fatcat:wiezcd7yn5cspmnjvc4qj6du7e

Leveraging Contact Forces for Learning to Grasp [article]

Hamza Merzic and Miroslav Bogdanovic and Daniel Kappler and Ludovic Righetti and Jeannette Bohg
2018 arXiv   pre-print
While visual feedback is important for inferring a grasp pose and reaching for an object, contact feedback offers valuable information during manipulation and grasp acquisition.  ...  While it is possible to learn grasping policies without contact sensing, our results suggest that contact feedback allows for a significant improvement of grasping robustness under object pose uncertainty  ...  Different from recent learning-based approaches for dexterous manipulation with multi-fingered hands [14, 24] , we assume feedback from contact sensors and noisy object pose estimates e.g. from visual  ... 
arXiv:1809.07004v1 fatcat:l2mh7doi5ndtxfegucct2cpt6m

Integrating vision, haptics and proprioception into a feedback controller for in-hand manipulation of unknown objects

Qiang Li, Christof Elbrechter, Robert Haschke, Helge Ritter
2013 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems  
We propose a feedback-based solution for the accurate manipulation of an unknown object in hand.  ...  Subsequently inverse hand kinematics is employed to retrieve joint-level motions, which are implemented on the robot with a position servo loop.  ...  Slobodan Vukanović for checking the English writing.  ... 
doi:10.1109/iros.2013.6696703 dblp:conf/iros/LiEHR13 fatcat:ojdc6e4webdsvpf3pt4jmtogpe

I-nteract: A cyber-physical system for real-time interaction with physical and virtual objects using mixed reality technologies for additive manufacturing [article]

Ammar Malik, Hugo Lhachemi, Robert Shorten
2020 arXiv   pre-print
The efficacy of the system has been demonstrated by generating 3D model using a novel scanning method to 3D print a customized orthopedic cast for human arm, by estimating spring rates of compression springs  ...  The system has been developed using mixed reality technologies such as HoloLens, for augmenting visual feedback, and haptic gloves, for augmenting haptic force feedback.  ...  The position sensors at the MCP joints of all the glove's digits are used to estimate the hand pose as well as position of the fingertips.  ... 
arXiv:2002.06280v1 fatcat:qnpdrj3fdret5datx2dt3aeb3q

Vision-driven Compliant Manipulation for Reliable, High-Precision Assembly Tasks [article]

Andrew S. Morgan, Bowen Wen, Junchi Liang, Abdeslam Boularias, Aaron M. Dollar, Kostas Bekris
2021 arXiv   pre-print
The proposed control method closes the loop through vision by tracking the relative 6D pose of objects in the relevant workspace.  ...  It adjusts the control reference of both the compliant manipulator and the hand to complete object insertion tasks via within-hand manipulation.  ...  Patel for his assistance in designing object-forming fingertips for the GR2 hand, in addition to the layered two-tone object geometries.  ... 
arXiv:2106.14070v1 fatcat:wscmrr7javcp3psrbgzj23kzay

Training a Feedback Loop for Hand Pose Estimation [article]

Markus Oberweger, Paul Wohlhart, Vincent Lepetit
2016 arXiv   pre-print
We propose an entirely data-driven approach to estimating the 3D pose of a hand given a depth image.  ...  We show that we can correct the mistakes made by a Convolutional Neural Network trained to predict an estimate of the 3D pose by using a feedback loop.  ...  Acknowledgements: This work was funded by the Christian Doppler Laboratory for Handheld Augmented Reality and the TU Graz FutureLabs fund.  ... 
arXiv:1609.09698v1 fatcat:izxrlmohznejldflee3s2x5im4

Training a Feedback Loop for Hand Pose Estimation

Markus Oberweger, Paul Wohlhart, Vincent Lepetit
2015 2015 IEEE International Conference on Computer Vision (ICCV)  
We propose an entirely data-driven approach to estimating the 3D pose of a hand given a depth image.  ...  We show that we can correct the mistakes made by a Convolutional Neural Network trained to predict an estimate of the 3D pose by using a feedback loop.  ...  Acknowledgements: This work was funded by the Christian Doppler Laboratory for Handheld Augmented Reality and the TU Graz FutureLabs fund.  ... 
doi:10.1109/iccv.2015.379 dblp:conf/iccv/OberwegerWL15 fatcat:qnthav3dcre5vjrisvhsiga7oa

I-nteract: A cyber-physical system for real-time interaction with physical and virtual objects using mixed reality technologies for additive manufacturing

Ammar Malik, Hugo Lhachemi, Robert Shorten
2020 IEEE Access  
The efficacy of the system has been demonstrated by generating a 3D model using a novel scanning method to 3D print a customized orthopedic cast for human arm, by estimating spring rates of compression  ...  The system has been developed using mixed reality technologies such as HoloLens, for augmenting visual feedback, and haptic gloves, for augmenting haptic force feedback.  ...  The position sensors at the MCP joints of all the glove's digits are used to estimate the hand pose as well as the position of the fingertips.  ... 
doi:10.1109/access.2020.2997533 fatcat:kbue652kbrf2npkwfkfllezjpa

PyMAF-X: Towards Well-aligned Full-body Model Regression from Monocular Images [article]

Hongwen Zhang, Yating Tian, Yuxiang Zhang, Mengcheng Li, Liang An, Zhenan Sun, Yebin Liu
2022 arXiv   pre-print
To address these issues, we propose a Pyramidal Mesh Alignment Feedback (PyMAF) loop in our regression network for well-aligned human mesh recovery and extend it as PyMAF-X for the recovery of expressive  ...  Moreover, when integrating part-specific estimations to the full-body model, existing solutions tend to either degrade the alignment or produce unnatural wrist poses.  ...  ACKNOWLEDGMENTS We would like to thank Xinchi Zhou, Wanli Ouyang, and Limin Wang for their help, feedback, and discussions in the early work of this paper.  ... 
arXiv:2207.06400v2 fatcat:oksezjiywza4zk2i4yyxbqqwza

Learning Haptic-based Object Pose Estimation for In-hand Manipulation with Underactuated Robotic Hands [article]

Osher Azulay, Inbar Ben-David, Avishai Sintov
2022 arXiv   pre-print
., kinesthetic and tactile sensing, for pose estimation and in-hand manipulation with underactuated hands.  ...  Consequently, pose estimation of a grasped object is usually performed based on visual perception.  ...  In addition, Table I presents accuracy results for the object pose estimation during in-hand manipulation.  ... 
arXiv:2207.02843v1 fatcat:trtaqp7hrrezzndkn6zqiazjly

Acquiring visual-motor models for precision manipulation with robot hands [chapter]

Martin Jägersand, Olac Fuentes, Randal Nelson
1996 Lecture Notes in Computer Science  
Dextrous high degree of freedom (DOF) robotic hands provide versatile motions for fine manipulation of potentially very different objects.  ...  We instead propose a combination of two techniques: the use of an approximate estimated motor model, based on the grasp tetrahedron acquired when grasping an object, and the use of visual feedback to achieve  ...  Error distributions for the 6 DOF positioning experiment are shown in fig. 5 . For open loop joint feedback, the object positioning errors are of two distinct types.  ... 
doi:10.1007/3-540-61123-1_174 fatcat:223d4e5xyvdvnkqu4wdiul2yci

A Framework for Force and Visual Control of Robot Manipulators [chapter]

Vincenzo Lippiello, Bruno Siciliano, Luigi Villani
2010 Springer Tracts in Advanced Robotics  
An algorithm for online estimation of the object pose is adopted, based on visual data as well as on force measurements. This information is used by a force/position controller.  ...  The resulting control scheme has a inner/outer structure where the outer loop performs pose estimation and the inner loop is devoted to interaction control.  ...  Dynamic Trajectory Planner Impedence Control Robot Target Object Pose Estimation Pose Control Desired Task Force feedback (1khz) Joint feedback (1khz) Pose feedback (26hz)  ... 
doi:10.1007/978-3-642-14743-2_31 fatcat:uuozs4p5evf4xi42m5lldr56ky

Model-based autonomous system for performing dexterous, human-level manipulation tasks

Nicolas Hudson, Jeremy Ma, Paul Hebert, Abhinandan Jain, Max Bajracharya, Thomas Allen, Rangoli Sharan, Matanya Horowitz, Calvin Kuo, Thomas Howard, Larry Matthies, Paul Backes (+1 others)
2013 Autonomous Robots  
Deliberate interaction with the environment is incorporated into planning and control strategies, which, when coupled with world estimation, allows for refinement of models and precise manipulation.  ...  The system takes advantage of sensory feedback immediately with little open-loop execution, attempting true autonomous reasoning and multi-step sequencing that adapts in the face of changing and uncertain  ...  Unlike most other systems, remote independent testing of the software was performed to ensure robustness and test for generality.  ... 
doi:10.1007/s10514-013-9371-y fatcat:zwlbjvof5vb2fksz4jzlzvik24
« Previous Showing results 1 — 15 out of 22,182 results