Pith. sign in

REVIEW 3 major objections 5 minor 6 cited by

DEXOP: A Device for Robotic Transfer of Dexterous Human Manipulation

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Perioperation—recording a human demonstration through a mechanically linked passive robot hand—produces robot-training data that beats teleoperation for equal collection time.

desk verdict Strong hardware and a credible throughput story, but the abstract's policy claim is not established: the only DEXOP condition also includes teleop demos and a teleop-only fine-tuning phase. read the letter →

arxiv 2509.04441 v2 pith:AEWKR3L2 submitted 2025-09-04 cs.RO cs.AIcs.CVcs.HC

Hao-Shu Fang , Branden Romero , Yichen Xie , Arthur Hu , Bo-Ruei Huang , Juan Alvarez , Matthew Kim , Gabriel Margolis
show 4 more authors
Kavya Anbarasu Masayoshi Tomizuka Edward Adelson Pulkit Agrawal
This is my paper · ORCID
classification cs.ROcs.AIcs.CVcs.HC
keywords perioperationdexterousmanipulationpassivehandexoskeletonwhole-handtactilesensingbehaviorcloningrobotdatacollectionteleoperation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Perioperation is a proposed third path for collecting robot-training data, distinct from simulation and teleoperation: a human wears a passive exoskeleton that is mechanically linked to a passive robot hand, so the person demonstrates the task with direct force feedback while the system records joint angles, wrist camera views, and whole-hand tactile images. The paper claims this makes demonstrations faster and more accurate than teleoperation, and that policies trained on DEXOP data perform better per minute of collection time: at roughly equal total time (139 versus 142 minutes), the DEXOP-mixed policy reaches a normalized six-stage success of 0.51 versus 0.36 for teleoperation alone. If true, DEXOP lowers the cost of the high-quality, contact-rich demonstrations that dexterous robot learning currently lacks. The practical stake is scalability: a human can gather robot-ready manipulation data in daily environments without operating a full robot.

What carries the argument

The central mechanism is the passive exoskeleton-hand pair: a wearable exoskeleton whose kinematic chain matches a passive robot hand, coupled by layered 4-bar linkages so each human finger joint drives the corresponding robot finger joint one-to-one, while contact forces on the robot hand travel back through the same linkages to the operator's hand. This force transparency, combined with mechanical pose mirroring, is what lets a person demonstrate contact-rich tasks quickly and accurately. The system also embeds camera-based whole-hand tactile sensors and a wrist camera, and the DEXOP-7 variant is co-designed with the deployment hand so the recorded examples align with the robot in kinematics, tactile sensing, and visual configuration.

What would settle it

Run the six-stage bulb-installation benchmark with a policy trained purely on an equal number of DEXOP demonstrations, with no teleoperation mix: if it scores at or below the teleoperation-only policy at matched demonstration count, the per-unit-time efficiency claim loses its support.

Watch

Extended reading notes

Core claim

The central claim is that mechanically coupling a human hand to a passive robot hand, rather than remotely driving a robot, produces demonstration data that transfers to a real robot hand and trains policies more efficiently than teleoperation. The headline result is a six-stage, bimanual bulb-installation task in which the policy trained on 160 DEXOP demonstrations plus 40 teleoperation demonstrations reaches a normalized cumulative success of 0.513, outperforming a 100-teleop policy matched for total collection time (0.355) and even a 200-teleop policy (0.425) trained on twice as many demonstrations and more than twice the collection time. The paper's proposed explanation is that operator proprioception through the linkage removes the visual-retargeting and force-blindness problems of teleoperation, so demonstrations are both faster and less biased; for example, teleoperators over-rotate the bulb while screwing because they cannot feel shear force, biasing the dataset away from task progression.

Load-bearing premise

The load-bearing premise is that the passive robot hand the operator wears matches the deployed robot hand closely enough in shape, motion, and tactile sensing that demonstrations transfer without conversion, and the paper itself relies on adding teleoperation data to absorb mismatches in the arm rather than testing DEXOP data alone.

Editorial extensions

If this is right

  • A robotics group can reach a given policy success rate in roughly a third of the operator time, because DEXOP collected the bulb-installation data about 2.7 times faster than teleoperation.
  • Perioperation data can be collected in natural environments without a full robot present, making large-scale dexterous data collection cheaper and more portable.
  • Whole-hand tactile and joint recordings, unlike joint-position-only data, retain enough contact information to recover joint torques via the Jacobian transpose, which is the paper's stated motivation for the sensor design.
  • The co-design method sets a template: if a hand and an exoskeleton are built together, demonstrations can transfer with minimal post-processing.
  • A small teleoperation correction set can absorb arm-exoskeleton calibration errors, suggesting a practical recipe of perioperation data plus a teleop supplement rather than pure teleoperation.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: the headline advantage is measured on a mixed dataset, so the paper does not yet isolate how much DEXOP alone contributes; if arm calibration were tightened, a pure-DEXOP comparison would make that quantity visible.
  • Editorial inference: the proprioceptive advantage is likely task-dependent; contact-rich stages such as screw tightening and flap folding should show the largest gap, whereas force-insensitive pick-and-place tasks may shrink it, so the per-unit-time claim should be read as task-specific.
  • Editorial inference: if whole-hand force recovery matures, DEXOP's tactile stream could support torque-level imitation, a form of supervision teleoperation data usually lacks, extending the method toward force-controlled assembly.
  • Editorial inference: the mechanical separation of human and robot hands suggests the design can be re-targeted to any hand whose kinematics are co-designed, making perioperation a general data-collection interface rather than a one-off device.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper introduces DEXOP, a passive hand exoskeleton for 'perioperation'—a proposed paradigm for collecting dexterous manipulation demonstrations from human operators while maximizing transfer to a robot hand. Three variants (DEXOP-12, -9, -7) are presented, with DEXOP-7 co-designed with the EyeSight Hand so that kinematics, tactile sensing, and visual configuration match. The paper evaluates hardware characteristics (force, workspace, speed), a four-participant user study comparing task throughput of DEXOP-7 against a teleoperation baseline and bare-hand upper bound, qualitative demonstrations of dexterous tasks, and policy-learning experiments on a bimanual bulb-installation task using ACT behavior cloning. The main quantitative claims are that DEXOP roughly doubles to triples data-collection throughput relative to teleoperation and that policies trained on a mixture of DEXOP and teleop demonstrations improve normalized cumulative success per unit data-collection time over teleop-only policies.

Significance. If the hardware and data-efficiency claims hold, DEXOP would be a useful contribution to the robot-learning data-collection toolbox: it is a novel, entirely passive device that couples a wearable human exoskeleton to a sensorized robot hand, and the co-design with EyeSight Hand is conceptually sound for closing the embodiment gap. The paper ships concrete hardware metrics, a project page, and a reproducible policy pipeline, and the authors explicitly state limitations (e.g., DEXOP data alone not sufficient). However, the central evidence for the headline claims is currently limited: the user study involves four participants with no statistical analysis, and the policy comparison is confounded by an added teleop fine-tuning stage in the DEXOP condition. The significance of the work would be substantially strengthened by addressing these two evidential gaps.

major comments (3)
  1. [§5.6, Table 2] The abstract's claim that 'policies learned with DEXOP data significantly improve task performance per unit time' is not supported by the reported experiment, because the condition labeled '160 DEXOP + 40 TeleOP' is not a policy learned from DEXOP data alone. Section 5.6 states that DEXOP-only training was not viable and that the policy was trained for 500 epochs on the mixed set followed by 300 epochs on the 40-demo TeleOP subset 'to help calibrate errors in the arm exoskeleton assembly'; the teleop-only baselines receive no analogous calibration fine-tuning. The 0.513 vs 0.355 normalized cumulative success comparison at comparable total collection time (139.3 vs 141.7 min) therefore conflates the effect of DEXOP data with the effect of the teleop subset, the extra fine-tuning schedule, and the larger total demonstration count (200 vs 100). A DEXOP-only condition and a '160 DEXOP + 40 TeleOP' condition without the subset fine-tuning (or an equivalent calibration fine-tune for the teleop-only conditions) are needed to isolate the contribution of DEXOP.
  2. [§4.2, Figure 6] The user study's throughput claims rest on only four participants with five trials per condition and no reported per-participant variance, confidence intervals, or significance tests. The paper reports average completions per minute (e.g., 6 vs 11 for drilling; 12 vs 22 for bottle opening) but does not show whether these differences are consistent across participants or driven by one or two individuals, nor whether task-order or practice effects were controlled. Given that the throughput advantage is a central theme of the paper, the study needs at least per-participant data, effect sizes, and a test (e.g., paired permutation or Wilcoxon) to support the claimed advantage.
  3. [§3.3 and §5.2] The paper motivates whole-hand tactile sensing as essential for capturing contact forces during manipulation, but the policy-learning experiment uses tactile images only from the distal phalanges (Section 5.2) and does not use the whole-hand tactile configuration emphasized in Section 3.3. Consequently, the evaluation does not test the claimed benefit of whole-hand tactile sensing for policy transfer; the paper's later discussion of force recovery as future work does not fully address this mismatch, since the design rationale is presented as a key advantage of DEXOP over prior data-collection devices.
minor comments (5)
  1. [§2.2] Typo: 'additional degrees of freedom aslo enable' should read 'also enable.'
  2. [§4] Typo: 'for these experimentss' should read 'for these experiments.'
  3. [Supplementary §S2] The sentence 'The implementation of DEXOP-9 closely follows the implementation of DEXOP-9 but just removes the ring finger' appears to have a typo; presumably DEXOP-9 follows DEXOP-12 while removing the ring finger.
  4. [Figure 6] The throughput bar chart shows only point estimates; adding per-participant points or error bars would make the variability in the four-participant study visible.
  5. [Table 2] The table reports standard errors for normalized cumulative success but no significance tests between conditions; given the overlapping error bars (0.513±0.032 vs 0.425±0.032), the claim of 'significantly improve' requires a formal test or explicit effect-size reporting.

Circularity Check

0 steps flagged · score 1.0 of 10

No circular derivation; central claims are empirical comparisons with only minor, non-load-bearing self-citation.

full rationale

The paper's contributions are empirical hardware/data-collection comparisons, not a derivation from inputs. The abstract's claim that policies learned with DEXOP data improve performance per unit time is supported by Table 2, which compares '160 DEXOP + 40 TeleOP' against 100/200 TeleOP baselines; this is an external empirical benchmark, so the claim does not reduce by construction to the inputs. The closest self-citation is the EyeSight Hand [29], the deployment hand co-designed with DEXOP-7 and the source of the tactile-sensing design. That citation is concrete engineering support documented in Section S3, not an unverified uniqueness theorem or smuggled ansatz, and the paper's main claims stand or fall on the hardware and policy experiments rather than on the citation alone. Section 5.6 concedes that 'DEXOP data alone may not be sufficient to deploy a learned policy,' and the headline policy condition mixes in 40 teleoperation demos plus a teleoperation-only fine-tuning stage; this is a real threat to the empirical claim's internal validity, but it is an experimental confound, not circularity in the derivation. No equation or fitted parameter is renamed as a prediction, and no central result is defined in terms of itself.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

No fitted numerical parameters were introduced; all results come from hardware design and experiments. The main assumptions are domain assumptions about kinematic equivalence, force transparency, and the sufficiency of tactile-plus-position data for policy learning, several of which the paper explicitly acknowledges in its limitations.

assumptions (5)
  • domain assumption The human hand kinematic chain can be approximated by DEXOP-12's 12 DoF (2-DoF MCP, PIP, 2-DoF TM, IP); omitted DIP and thumb MCP joints do not prevent natural demonstration of the evaluated tasks.
    Section 3.1 states these joints were chosen to maximize task diversity; the user study and policy experiments rely on this coverage.
  • domain assumption The mechanical 4-bar linkage transmits human finger motion to the passive robot hand and returns interaction forces without significant backlash, slip, or delay, providing proprioceptive feedback.
    Sections 3.2 and 4.2; the reported throughput gains depend on force transparency and kinematic coupling.
  • domain assumption The co-designed DEXOP-7 and EyeSight Hand have matching kinematics, sensors, and visual configuration, so demonstrations transfer without retargeting or further data processing.
    Section 5.3 states 'no further data processing is needed to close the embodiment gap.' The paper also concedes arm-level mismatch required adding teleop data in Section 5.6, making hand-level equivalence load-bearing.
  • domain assumption Behavior cloning with ACT and the specified augmentations is a valid way to measure the utility of demonstration data.
    Section 5.5; this is a standard learning method, not derived or verified within the paper.
  • domain assumption Whole-hand tactile data plus joint positions are sufficient for learning contact-rich policies without direct joint torque measurement.
    Section 3.3 says torque cannot be captured directly and force recovery is left to future work; the policy learning experiment uses only distal tactile images and joint angles.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DEXOP: A Device for Robotic Transfer of Dexterous Human Manipulation." pith.science (2026). https://pith.science/paper/AEWKR3L2

@misc{pith2026250904441,
  author       = {Pith},
  title        = {Pith review of: DEXOP: A Device for Robotic Transfer of Dexterous Human Manipulation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AEWKR3L2}},
  note         = {Machine review of arXiv:2509.04441}
}
read the original abstract

We introduce perioperation, a paradigm for robotic data collection that sensorizes and records human manipulation while maximizing the transferability of the data to real robots. We implement this paradigm in DEXOP, a passive hand exoskeleton designed to maximize human ability to collect rich sensory (vision + tactile) data for diverse dexterous manipulation tasks in natural environments. DEXOP mechanically connects human fingers to robot fingers, providing users with direct contact feedback (via proprioception) and mirrors the human hand pose to the passive robot hand to maximize the transfer of demonstrated skills to the robot. The force feedback and pose mirroring make task demonstrations more natural for humans compared to teleoperation, increasing both speed and accuracy. We evaluate DEXOP across a range of dexterous, contact-rich tasks, demonstrating its ability to collect high-quality demonstration data at scale. Policies learned with DEXOP data significantly improve task performance per unit time of data collection compared to teleoperation, making DEXOP a powerful tool for advancing robot dexterity. Our project page is at https://dex-op.github.io.

Figures

Figures reproduced from arXiv: 2509.04441 by the authors.

Figure 1
Figure 1. (a) DEXOP is a passive exoskeleton that links human hand movements to passive robotic hand move￾ments through mechanical linkages. (b) DEXOP enables humans to collect task demonstrations of diverse and highly dexterous tasks. (c) Data collected by DEXOP can be used to train policies that transfer to robots. Abstract: We introduce perioperation, a paradigm for robotic data collection that sensorizes and records human… view at source ↗
Figure 2
Figure 2. Hardware overview: (a) Variants of DEXOP: DEXOP-12 (4 fingers and 12 DOF), the most advanced DEXOP; DEXOP-9 without the ring finger (3 fingers and 9 DOF); and DEXOP-7 without abduction joints on index and middle fingers (3 fingers, 7 DOF, and co-designed with EyeSight hand). (b) Illustration of a human hand controlling the DEXOP-9 to grasp a ball, along with the tactile sensor readings and robotic hand pose. the two… view at source ↗
Figure 3
Figure 3. A depiction of joints in a human hand’s kinematic chain. The blue joints are presented in DEXOP-12. In this work, we present three DEXOP variants: DEXOP￾12 with 4 fingers and 12 degrees of freedom, DEXOP-9 with 3 fingers and 9 degrees of freedom, and DEXOP￾7 with 3 fingers and 7 degrees of freedom. The first two are used to demonstrate the dexterous perioperation, while the third is co-designed with a robotic hand [… view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: (a) Exploded view of the DEXOP-12 system. (b) Top: Annotated view of the 4-bar linkages coupling the index, middle and ring fingers of the robot hand and exoskeleton. Bottom: Annotated view of the rotary linkage system coupling the thumb of the robot hand and exoskelet…
Figure 5
Figure 5. Figure 5: Illustration of evaluation tasks. Drilling: the user must pick up a drill standing upright on a table, then insert the drill bit into an M2 screw head and tighten it by actuating the drill. Bottle opening: with the bottle placed within the workspace of the hand, the us…
Figure 6
Figure 6. Figure 6: Comparison of task throughput of the drilling, bulb installation, box packaging and bottle opening [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]
Figure 7
Figure 7. Figure 7: Various tasks illustrating the dexterity that [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]
Figure 8
Figure 8. Figure 8: (a) A Unitree H1 with EyeSight hands used for policy evaluation. (b) The [PITH_FULL_IMAGE:figures/full_fig_p012_8.png]
Figure 9
Figure 9. Figure 9: Illustration of our task for policy learning. Wrist camera images highlighted in green showcase [PITH_FULL_IMAGE:figures/full_fig_p013_9.png]
Figure 10
Figure 10. Figure 10: (a) Success rates of policies trained on mixed [PITH_FULL_IMAGE:figures/full_fig_p014_10.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. High Fidelity Capture, Reconstruction, and Transfer of Human Demonstrations for Robot-Assisted Bathing

    cs.RO 2026-08 conditional novelty 6.0 of 10

    A contact-centered pipeline yields the first multimodal dataset of clinician-performed bathing and transfers the demonstrations to a soft robotic hand and arm.

  2. DexDirect: Direct Kinesthetic Arm Guidance for Efficient Dexterous Demonstration Collection

    cs.RO 2026-07 conditional novelty 6.0 of 10

    A hybrid kinesthetic-arm-plus-webcam-hand teleoperation interface achieved 17x/3x higher demonstration throughput than vision baselines and trained a 90%-success pick-and-place policy in a ten-person study.

  3. Asynchronous Multimodal Diffusion Policy Composition via Latency-Aware Guidance Fusion

    cs.RO 2026-07 conditional novelty 6.0 of 10

    LAG-Fusion composes asynchronous diffusion policies by rebasing delayed guidance into the current action frame and fusing with latency-aware weights, improving contact-rich manipulation performance.

  4. Towards Human-level Dexterous Teleoperation

    cs.RO 2026-07 conditional novelty 6.0 of 10

    A single-stage RL co-tracking controller trained on consecutive human-derived hand–object subgoals achieves ~75% real-robot success on long-horizon dexterous teleoperation where baselines fail.

  5. Play2Perfect: What Matters in Dexterous Play Pretraining for Precise Assembly?

    cs.RO 2026-06 unverdicted novelty 6.0 of 10

    Task-agnostic RL play pretraining on diverse objects yields a reusable dexterous prior that makes sparse-reward assembly learning ~33× more sample-efficient and enables zero-shot sim-to-real transfer on tight insertio...

  6. AnyDexRT: Calibration-Free Dexterous Hand Retargeting with Few-Shot Human Guidance

    cs.RO 2026-07 conditional novelty 5.5 of 10

    Self-supervised fingertip mapping with few-shot human anchors and a pinch contact classifier yields calibration-free, more intuitive retargeting across diverse human-like robot hands.

Reference graph

Works this paper leans on

57 extracted references · 29 canonical work pages · cited by 6 Pith papers

  1. [1]

    T. Chen, J. Xu, and P. Agrawal. A system for general in-hand object re-orientation. In Conference on Robot Learning, pages 297–307. PMLR, 2022

    work page 2022

  2. [2]

    T. Chen, M. Tippur, S. Wu, V . Kumar, E. Adelson, and P. Agrawal. Visual dexterity: In-hand reorientation of novel and complex object shapes. Science Robotics, 8(84):eadc9244, 2023

    work page 2023

  3. [3]

    H. Qi, B. Yi, S. Suresh, M. Lambeta, Y . Ma, R. Calandra, and J. Malik. General in-hand object rotation with vision and touch. In Conference on Robot Learning, pages 2549–2564. PMLR, 2023

    work page 2023

  4. [4]

    O. M. Andrychowicz, B. Baker, M. Chociej, R. Jozefowicz, B. McGrew, J. Pachocki, A. Petron, M. Plap- pert, G. Powell, A. Ray, et al. Learning dexterous in-hand manipulation. The International Journal of Robotics Research, 39(1):3–20, 2020

    work page 2020

  5. [5]

    Akkaya, M

    I. Akkaya, M. Andrychowicz, M. Chociej, M. Litwin, B. McGrew, A. Petron, A. Paino, M. Plappert, G. Powell, R. Ribas, et al. Solving rubik’s cube with a robot hand. arXiv preprint arXiv:1910.07113, 2019

    arXiv 1910

  6. [6]

    H. Zhu, A. Gupta, A. Rajeswaran, S. Levine, and V . Kumar. Dexterous manipulation with deep rein- forcement learning: Efficient, general, and low-cost. In 2019 International Conference on Robotics and Automation (ICRA), pages 3651–3657. IEEE, 2019

    2019

  7. [7]

    Torne, A

    M. Torne, A. Simeonov, Z. Li, A. Chan, T. Chen, A. Gupta, and P. Agrawal. Reconciling reality through simulation: A real-to-sim-to-real approach for robust manipulation. arXiv preprint arXiv:2403.03949 , 2024

    arXiv 2024

  8. [8]

    L. L. Ankile, A. Simeonov, I. Shenfeld, M. T. Villasevil, and P. Agrawal. From imitation to refinement– residual rl for precise visual assembly. In CoRL 2024 Workshop on Mastering Robot Manipulation in a World of Abundant Data

    work page 2024

Show all 57 references
  1. [9]

    Y . Park, J. S. Bhatia, L. Ankile, and P. Agrawal. Dexhub and dart: Towards internet scale robot data collection. arXiv preprint arXiv:2411.02214, 2024

    2024 arXiv

  2. [10]

    Huang, I

    W. Huang, I. Mordatch, P. Abbeel, and D. Pathak. Generalization in dexterous manipulation via geometry- aware multi-task learning. arXiv preprint arXiv:2111.03062, 2021

    2021 arXiv

  3. [11]

    Y . Park, G. B. Margolis, and P. Agrawal. Position: Automatic environment shaping is the next frontier in rl. In Forty-first International Conference on Machine Learning

  4. [12]

    Antotsiou, G

    D. Antotsiou, G. Garcia-Hernando, and T.-K. Kim. Task-oriented hand motion retargeting for dexter- ous manipulation imitation. In Proceedings of the European Conference on Computer Vision (ECCV) Workshops, pages 0–0, 2018

    2018

  5. [13]

    Orbik, S

    J. Orbik, S. Li, and D. Lee. Human hand motion retargeting for dexterous robotic hand. In 2021 18th International Conference on Ubiquitous Robots (UR), pages 264–270. IEEE, 2021

    2021

  6. [14]

    Qin, Y .-H

    Y . Qin, Y .-H. Wu, S. Liu, H. Jiang, R. Yang, Y . Fu, and X. Wang. Dexmv: Imitation learning for dexterous manipulation from human videos. arXiv preprint arXiv:2108.05877, 2021. 17

    2021 arXiv

  7. [15]

    K. Shaw, S. Bahl, and D. Pathak. Videodex: Learning dexterity from internet videos. In Conference on Robot Learning, pages 654–665. PMLR, 2023

    2023

  8. [16]

    Mandikal and K

    P. Mandikal and K. Grauman. Dexvip: Learning dexterous grasping with human hand pose priors from video. In Conference on Robot Learning, pages 651–661. PMLR, 2022

    2022

  9. [17]

    Radosavovic, X

    I. Radosavovic, X. Wang, L. Pinto, and J. Malik. State-only imitation learning for dexterous manipulation. In 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 7865–7871. IEEE, 2021

    2021

  10. [18]

    C. Wang, H. Shi, W. Wang, R. Zhang, L. Fei-Fei, and C. K. Liu. Dexcap: Scalable and portable mocap data collection system for dexterous manipulation. arXiv preprint arXiv:2403.07788, 2024

    2024 arXiv

  11. [19]

    Y . Qin, H. Su, and X. Wang. From one hand to multiple hands: Imitation learning for dexterous manipu- lation from single-camera teleoperation. arXiv preprint arXiv:2204.12490, 2022

    2022 arXiv

  12. [20]

    S. Yang, M. Liu, Y . Qin, R. Ding, J. Li, X. Cheng, R. Yang, S. Yi, and X. Wang. Ace: A cross-platform visual-exoskeletons system for low-cost dexterous teleoperation.arXiv preprint arXiv:2408.11805, 2024

    2024 arXiv

  13. [21]

    R. Ding, Y . Qin, J. Zhu, C. Jia, S. Yang, R. Yang, X. Qi, and X. Wang. Bunny-visionpro: Real-time bimanual dexterous teleoperation for imitation learning. arXiv preprint arXiv:2407.03162, 2024

    2024 arXiv

  14. [22]

    Cheng, J

    X. Cheng, J. Li, S. Yang, G. Yang, and X. Wang. Open-television: teleoperation with immersive active visual feedback. arXiv preprint arXiv:2407.01512, 2024

    2024 arXiv

  15. [23]

    Sarakoglou, A

    I. Sarakoglou, A. Brygo, D. Mazzanti, N. G. Hernandez, D. G. Caldwell, and N. G. Tsagarakis. Hexotrac: A highly under-actuated hand exoskeleton for finger tracking and force feedback. In 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 103...

    2016

  16. [24]

    Zhang, S

    H. Zhang, S. Hu, Z. Yuan, and H. Xu. Doglove: Dexterous manipulation with a low-cost open-source haptic force feedback glove. arXiv preprint arXiv:2502.07730, 2025

    2025 arXiv

  17. [25]

    Sundaram, P

    S. Sundaram, P. Kellnhofer, Y . Li, J.-Y . Zhu, A. Torralba, and W. Matusik. Learning the signatures of the human grasp using a scalable tactile glove. Nature, 569(7758), 2019. doi:10.1038/s41586-019-1234-z

    2019 doi

  18. [26]

    Y . Luo, C. Liu, Y . J. Lee, J. DelPreto, K. Wu, M. Foshey, D. Rus, T. Palacios, Y . Li, A. Torralba, et al. Adaptive tactile interaction transfer via digitally embroidered smart gloves. Nature communications, 15 (1):868, 2024

    2024

  19. [27]

    C. Xing, H. Li, Y .-L. Wei, T.-A. Ren, T. Tu, Y . Lin, E. Schumann, W.-S. Zheng, and M. R. Cutkosky. Taccap: A wearable fbg-based tactile sensor for seamless human-to-robot skill transfer. arXiv e-prints, pages arXiv–2503, 2025

    2025

  20. [28]

    M. Xu, H. Zhang, Y . Hou, Z. Xu, L. Fan, M. Veloso, and S. Song. Dexumi: Using human hand as the universal manipulation interface for dexterous manipulation. arXiv preprint arXiv:2505.21864, 2025

    2025

  21. [29]

    Romero, H.-S

    B. Romero, H.-S. Fang, P. Agrawal, and E. Adelson. Eyesight hand: Design of a fully-actuated dex- terous robot hand with integrated vision-based tactile sensors and compliant actuation. arXiv preprint arXiv:2408.06265, 2024

    2024 arXiv

  22. [30]

    W. K. Do, A. K. Dhawan, M. Kitzmann, and M. K. III. Densetact-mini: An optical tactile sensor for grasping multi-scale objects from flat surfaces. 2023. arXiv preprint arXiv:2309.08860

    2023 arXiv

  23. [31]

    Laron, E

    A. Laron, E. Sne, Y . Perets, and A. Sintov. Print-n-grip: A disposable, compliant, scalable and one-shot 3d-printed multi-fingered robotic hand. 2024. arXiv preprint arXiv:2401.16463

    2024 arXiv

  24. [32]

    Burgess and E

    M. Burgess and E. H. Adelson. Grasp everything (get): 1-dof, 3-fingered gripper with tactile sensing for robust grasping. 2025. arXiv preprint arXiv:2505.09771

    2025 arXiv

  25. [33]

    K. Shaw, A. Agarwal, and D. Pathak. Leap hand: Low-cost, efficient, and anthropomorphic hand for robot learning. arXiv preprint arXiv:2309.06440, 2023

    2023 arXiv

  26. [34]

    Martinez-Hernandez, M

    U. Martinez-Hernandez, M. Szollosy, L. W. Boorman, H. Kerdegari, and T. J. Prescott. Towards a wear- able interface for immersive telepresence in robotics. In Interactivity, Game Creation, Design, Learning, and Innovation: 5th International Conference, ArtsIT 2016, and First I...

    2016

  27. [35]

    Z.-H. Yin, C. Wang, L. Pineda, F. Hogan, K. Bodduluri, A. Sharma, P. Lancaster, I. Prasad, M. Kalakr- ishnan, J. Malik, et al. Dexteritygen: Foundation controller for unprecedented dexterity. arXiv preprint arXiv:2502.04307, 2025. 18

    2025 arXiv

  28. [36]

    R. Luo, C. Wang, C. Keil, D. Nguyen, H. Mayne, S. Alt, E. Schwarm, E. Mendoza, T. Padır, and J. P. Whitney. Team northeastern’s approach to ana xprize avatar final testing: A holistic approach to telepres- ence and lessons learned. In 2023 IEEE/RSJ International Conference on ...

    2023

  29. [37]

    T. Z. Zhao, V . Kumar, S. Levine, and C. Finn. Learning fine-grained bimanual manipulation with low-cost hardware. arXiv preprint arXiv:2304.13705, 2023

    2023 arXiv

  30. [38]

    P. Wu, Y . Shentu, Z. Yi, X. Lin, and P. Abbeel. Gello: A general, low-cost, and intuitive teleoperation framework for robot manipulators. arXiv preprint arXiv:2309.13037, 2023

    2023 arXiv

  31. [39]

    Fang, H.-S

    H. Fang, H.-S. Fang, Y . Wang, J. Ren, J. Chen, R. Zhang, W. Wang, and C. Lu. Airexo: Low-cost exoskeletons for learning whole-arm manipulation in the wild. In 2024 IEEE International Conference on Robotics and Automation (ICRA), pages 15031–15038. IEEE, 2024

    2024

  32. [40]

    C. Chi, Z. Xu, C. Pan, E. Cousineau, B. Burchfiel, S. Feng, R. Tedrake, and S. Song. Universal manipu- lation interface: In-the-wild robot teaching without in-the-wild robots. arXiv preprint arXiv:2402.10329, 2024

    2024 arXiv

  33. [41]

    S. Song, A. Zeng, J. Lee, and T. Funkhouser. Grasping in the wild: Learning 6dof closed-loop grasping from low-cost demonstrations. IEEE Robotics and Automation Letters, 5(3):4978–4985, 2020

    2020

  34. [42]

    N. M. M. Shafiullah, A. Rai, H. Etukuru, Y . Liu, I. Misra, S. Chintala, and L. Pinto. On bringing robots home. arXiv preprint arXiv:2311.16098, 2023

    2023 arXiv

  35. [43]

    T. Tao, M. K. Srirama, J. J. Liu, K. Shaw, and D. Pathak. Dexwild: Dexterous human interactions for in-the-wild robot policies. arXiv preprint arXiv:2505.07813, 2025

    2025 arXiv

  36. [44]

    Z. Si, J. E. Chen, M. E. Karagozler, A. Bronars, J. Hutchinson, T. Lampe, N. Gileadi, T. Howell, S. Sal- iceti, L. Barczyk, et al. Exostart: Efficient learning for dexterous manipulation with sensorized exoskele- ton demonstrations. arXiv preprint arXiv:2506.11775, 2025

    2025

  37. [45]

    S.-S. Yun, B. B. Kang, and K.-J. Cho. Exo-glove pm: An easily customizable modularized pneumatic assistive glove. IEEE Robotics and Automation Letters, 2(3):1725–1732, 2017

    2017

  38. [46]

    B. B. Kang, H. Choi, H. Lee, and K.-J. Cho. Exo-glove poly ii: A polymer-based soft wearable robot for the hand with a tendon-driven actuation system. Soft robotics, 6(2):214–227, 2019

    2019

  39. [47]

    Y . Yun, P. Agarwal, J. Fox, K. E. Madden, and A. D. Deshpande. Accurate torque control of finger joints with ut hand exoskeleton through bowden cable sea. In 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 390–397. IEEE, 2016

    2016

  40. [48]

    Agarwal, J

    P. Agarwal, J. Fox, Y . Yun, M. K. O’Malley, and A. D. Deshpande. An index finger exoskeleton with series elastic actuation for rehabilitation: Design, control and performance characterization. The International Journal of Robotics Research, 34(14):1747–1772, 2015

    2015

  41. [49]

    P. W. Ferguson, Y . Shen, and J. Rosen. Hand exoskeleton systems—overview.Wearable Robotics, pages 149–175, 2020

    2020

  42. [50]

    W. P. Cooney, M. J. Lucca, E. Y . S. Chao, and R. L. Linscheid. The kinesiology of the thumb trapeziometacarpal joint. The Journal of Bone and Joint Surgery. American Volume , 63(9):1371–1381,

  43. [51]

    C. Chen, Z. Yu, H. Choi, M. Cutkosky, and J. Bohg. Dexforce: Extracting force-informed actions from kinesthetic demonstrations for dexterous manipulation. IEEE Robotics and Automation Letters, 2025

    2025

  44. [52]

    H. Fang, C. Wang, Y . Wang, J. Chen, S. Xia, J. Lv, Z. He, X. Yi, Y . Guo, X. Zhan, L. Yang, W. Wang, C. Lu, and H.-S. Fang. Airexo-2: Scaling up generalizable robotic imitation learning with low-cost exoskeletons. arXiv preprint arXiv:, 2025

    2025

  45. [53]

    K. An, L. Askew, and E. Chao. Biomechanics and functional assessment of upper extremities. InTrends in ergonomics/human factors III, volume 1986, pages 573–580. Elsevier Science Publishers North-Holland, 1986

    1986

  46. [54]

    J. R. Napier. The prehensile movements of the human hand. The Journal of Bone and Joint Surgery. British Volume, 38-B(4):902–913, 1956. doi:10.1302/0301-620X.38B4.902

    1956 doi

  47. [55]

    M. W. Marzke, K. L. Wullstein, and S. F. Viegas. Evolution of the power (“squeeze”) grip and its mor- phological correlates in hominids. American Journal of Physical Anthropology , 89(3):283–298, 1992. doi:10.1002/ajpa.1330890303

    1992 doi

  48. [56]

    M. Heo, Y . Lee, D. Lee, and J. J. Lim. Furniturebench: Reproducible real-world benchmark for long- horizon complex manipulation. In Robotics: Science and Systems, 2023. 19 Supplementary S1. Electronics of DEXOP: Each revolute joint in the passive robotic hand is equipped with...

    2023

  49. [1981]

    doi:10.2106/00004623-198163090-00004

    doi

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.