Pith. sign in

REVIEW 3 major objections 5 minor 25 references

HumanoidUMI: Bridging Robot-Free Demonstrations and Humanoid Whole-Body Manipulation

T0 review · 3 major / 5 minor · reviewed 2026-07-12 · grok-4.5

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

Pith's one-line read Portable VR and handheld grippers let people teach humanoids whole-body skills without ever touching the robot.

desk verdict Solid portable robot-free humanoid demo pipeline with real G1 results and a clear efficiency win; the transfer claim is under-supported without a teleop-trained policy baseline. read the letter →

arxiv 2606.27239 v2 pith:5YK7W5VP submitted 2026-06-25 cs.RO

Hongwu Wang , Chenhao Yu , Youhao Hu , Jiachen Zhang , Yuanyuan Li , Shaqi Luo This is my paper
classification cs.RO
keywords humanoidrobotrobot-freedatacollectionwhole-bodymanipulationUniversalInterfacespatialkeypointretargetingdiffusionpolicyvisuomotorlearningloco-manipulation
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

Humanoid robots need coordinated whole-body demos—grasping while bending, walking while handing something over—but collecting them by teleoperating the robot is slow, requires access to hardware, and demands skilled operators. HumanoidUMI instead records natural human motion with a lightweight VR headset, a few body trackers, and UMI-style handheld grippers that also capture wrist-view video and gripper width. The system stores only sparse task-relevant keypoints (pelvis, two tool-center points, two feet, optionally knees), trains a diffusion policy to predict future keypoint motion from the wrist views and lower-body proprioception, then uses Spatial Keypoint Retargeting to turn those predictions into robot-native root and joint references that a learned whole-body controller can track. On a Unitree G1 the pipeline produces working policies for cluttered pick-and-place, bimanual collection, timed ball throwing, under-table waste disposal, and walking coffee delivery, while gathering valid demonstrations roughly twice as fast as a strong teleoperation baseline—and far faster for locomotion-plus-manipulation.

What carries the argument

Spatial Keypoint Retargeting (SKR): a morphology-aware bridge that keeps metric spatial relationships among five (or seven) task-space keypoints, applies only a fixed anisotropic vertical scale (λ_leg = 0.75) to leg-related points to match robot height, then solves a two-stage weighted inverse-kinematics problem to produce executable root pose and joint references.

What would settle it

On the same Unitree G1 and the same five tasks, replace SKR with a naïve global rescaling or pure joint-space retargeting of the identical human demos and measure whether success rates collapse relative to the reported SKR results, especially on under-table waste disposal and walking coffee delivery.

Watch

Extended reading notes

Core claim

Robot-free human demonstrations collected with a portable VR–UMI interface can be turned into deployable humanoid whole-body skills: a high-level diffusion policy predicts sparse keypoint trajectories, Spatial Keypoint Retargeting converts them into feasible robot references, and a learned whole-body controller executes them with balance on a physical Unitree G1 across five real tasks spanning single-arm, bimanual, dynamic, bending, and loco-manipulation behaviors.

Load-bearing premise

A handful of body keypoints plus wrist cameras, after one fixed leg-length scale and weighted inverse kinematics, still carries enough geometric information to close the human-to-robot body gap for coordinated whole-body skills.

Editorial extensions

If this is right

  • Valid whole-body training data can be collected without the target humanoid present, lowering hardware and safety barriers.
  • Novice operators can produce usable loco-manipulation demos at rates close to experienced users, reducing skill dependence.
  • A single sparse keypoint interface plus a general whole-body controller can cover both quasi-static manipulation and timing-sensitive dynamic release.
  • Hierarchical separation of task-space prediction, retargeting, and low-level tracking makes the same human demos reusable across different humanoid morphologies once SKR is re-calibrated.

Reading between the lines

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

  • If the sparse keypoint set proves sufficient, large-scale in-the-wild humanoid datasets could be gathered by ordinary people with consumer VR kits rather than robot labs.
  • The same hierarchy may transfer to other morphologically mismatched platforms (e.g., different bipedal robots) by changing only the SKR scale and IK weights.
  • Wrist-view-only sensing may limit long-horizon spatial memory; adding a sparse third-person or head-mounted stream is a natural next stress test.
Share X Bluesky LinkedIn Reddit HN

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. HumanoidUMI presents a robot-free pipeline for collecting and deploying humanoid whole-body manipulation demonstrations. Using a PICO VR full-body capture setup and UMI-inspired dual handheld grippers, the system records wrist-view fisheye images, gripper widths, and a sparse five-keypoint (or seven-keypoint) trajectory of pelvis, TCPs, feet (and optionally knees). A diffusion policy predicts future relative keypoint motions and gripper actions conditioned on wrist views and lower-body proprioception; Spatial Keypoint Retargeting (SKR) applies a fixed anisotropic leg scale λ_leg=0.75 and weighted two-stage IK to produce robot-native root/joint references; a learned whole-body controller tracks those references on a Unitree G1. Real-robot experiments cover five tasks (cluttered pick-and-place, bimanual vegetable collection, dynamic ball throwing, under-table waste disposal, walking coffee delivery), with ablations of SKR vs GMR, latency matching, and 5- vs 7-keypoint layouts, plus a throughput comparison against TWIST2 teleoperation.

Significance. If the transfer claim holds, the work offers a practical, low-cost route to scalable humanoid whole-body data without robot-in-the-loop teleoperation, extending the UMI paradigm to coordinated loco-manipulation. Strengths include a clear hierarchical separation of task-space prediction, morphology-aware retargeting, and low-level tracking; real G1 deployment across five qualitatively different tasks; and concrete ablations (SKR, latency matching, knee keypoints) plus a quantified collection-efficiency gain over TWIST2. The sparse keypoint interface and online SKR visualization during collection are useful engineering contributions for the community.

major comments (3)
  1. The central claim is that robot-free demos yield transferable, deployable whole-body skills, not only that demos can be collected faster. §IV.A–B report success rates only for HumanoidUMI-trained policies (and internal ablations of SKR, latency matching, and 5- vs 7-keypoint layouts). §IV.C / Table I compare only valid-demo throughput vs TWIST2, not policy quality under matched data budgets or identical evaluation protocols. Without a teleop-trained (or embodiment-matched) policy baseline on the same five tasks, high success is consistent with either successful morphology-gap closure or with tasks that a competent whole-body controller can solve once any reasonable high-level policy is available. An apples-to-apples policy comparison is load-bearing for the transfer claim.
  2. §III.C, Eqs. (4)–(5): SKR relies on a single free scale λ_leg=0.75 and fixed two-stage weighted IK. The paper does not report sensitivity of success rates to λ_leg, to the IK weights w_i / λ_q, or to alternative morphology-compensation schemes. Given that the weakest technical assumption is that this sparse anisotropic adjustment plus IK sufficiently closes the human–G1 gap for loco-manipulation, a short sensitivity study (or failure-mode analysis when λ_leg is misspecified) is needed to support that the interface is robust rather than tuned to the present demonstrator and robot.
  3. Success rates in Figs. 6–7 are given as point estimates over 20 trials with no error bars, confidence intervals, or statistical comparison. For the ablations that are used to justify SKR and latency matching, binomial CIs or a simple significance test would make the claimed improvements interpretable and would strengthen the empirical support for the hierarchical design.
minor comments (5)
  1. Fig. 6 and Fig. 7 captions refer to “left-side bar plots” of success rates; ensure the published figures include numerical labels or a table of exact rates so readers can recover the values without visual estimation.
  2. §III.B: the action dimension is stated as 5×9+2=47 (and 65 for seven keypoints). A brief note clarifying that the 6-D rotation is the continuous representation of Zhou et al. and how it is recovered at inference would help reproducibility.
  3. Latency-matching procedure is only referenced to UMI; a short description of the measured delays (camera, gripper encoder, policy, control) for the G1 setup would make the dynamic-throwing result more self-contained.
  4. Related work correctly positions HuMI and HoMMI; a one-sentence clarification of how online SKR during collection differs from HuMI’s retargeting coupling would sharpen the novelty claim.
  5. Minor notation: T_rel and T_pel in Eqs. (1)–(2) would benefit from an explicit statement that all poses are SE(3) and that gripper widths remain absolute scalars after min–max normalization of translations.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: empirical systems paper whose success rates and throughput are external physical measurements, not algebraic restatements of fitted inputs.

full rationale

HumanoidUMI is a robot-free data-collection and hierarchical control pipeline (VR–UMI demos → diffusion keypoint policy → Spatial Keypoint Retargeting → learned whole-body controller). The load-bearing claims are (i) physical success rates on five Unitree G1 tasks and (ii) valid-demo throughput vs TWIST2 (Table I). Neither quantity is defined in terms of the other, nor recovered by construction from a fitted constant. λ_leg=0.75 (Eq. 4) is an explicit free morphology scale, not a predicted observable; SKR (Eq. 5) is a constrained IK design choice, not a uniqueness theorem. Self-citations to UMI/HuMI/OmniUMI supply background methods and do not define the reported success rates or force the efficiency comparison. Policy “predictions” are standard learned action chunks evaluated on real-robot trials. No self-definitional loop, fitted-input-as-prediction, load-bearing self-citation uniqueness claim, or renamed known result appears in the derivation chain. Circularity score is therefore 0; concerns about missing teleop-trained policy baselines are completeness/correctness issues, not circularity.

Assumptions & free parameters 4 free parameters · 5 assumptions · 3 invented entities

The central claim rests on engineering choices and domain assumptions rather than new physics: sparse keypoints suffice, a fixed leg scale closes morphology gap, diffusion policy + learned WBC can track retargeted references, and 20-trial success rates generalize. Free parameters are few but load-bearing (especially λ_leg). Invented entities are system modules (HumanoidUMI, SKR, keypoint interface), not new physical objects; independent evidence is the robot experiments themselves, which are not externally replicated here.

free parameters (4)
  • λ_leg (pelvis-relative vertical leg scale)
    Hand-set to 0.75 to compensate human–G1 leg length (§III.C, Eq. 4). Directly affects crouch/stand feasibility and is not learned from data.
  • Diffusion action horizon H
    Fixed H=48 for keypoint/gripper chunk prediction (§III.B); affects temporal credit and latency matching for throwing.
  • SKR IK weights w_i, λ_q and two-stage prioritization
    Weighted constrained IK (Eq. 5) with staged foot-then-pelvis/TCP priorities; values not fully specified but control retarget feasibility.
  • Low-level residual action scale s and clip a_max
    Controller residual joint-position scaling (Eq. 7) chosen for stable PD tracking on G1.
assumptions (5)
  • domain assumption Sparse task-space keypoints (pelvis, TCPs, feet; optional knees) plus wrist RGB and lower-body proprioception are a sufficient interface for whole-body visuomotor intent.
    Core modeling choice in §III.A–B; underpins both data format and high-level policy action space.
  • ad hoc to paper Morphology mismatch is adequately handled by anisotropic vertical leg scaling plus constrained IK without full mesh/contact retargeting.
    SKR design (§III.C); alternative global rescaling (GMR) is ablated but the scale form is paper-specific.
  • domain assumption A learned whole-body motion tracker trained in simulation (MJLab) can execute SKR references on the physical G1 with balance under contact tasks.
    §III.D; required for ‘deployable’ claim; not re-proven here beyond task success.
  • domain assumption Diffusion Policy in relative pelvis-frame keypoint space is an appropriate high-level learner for these demos.
    §III.B cites Chi et al.; architecture choice assumed workable after standard conditioning with DINOv2.
  • domain assumption Valid demonstration throughput in 10 minutes is a meaningful proxy for collection efficiency relevant to scalable humanoid learning.
    §IV.C Table I; does not measure final policy quality parity with teleop data.
invented entities (3)
  • HumanoidUMI data-collection stack (PICO full-body + UMI-inspired dual grippers)
    purpose: Robot-free capture of synchronized keypoints, wrist views, and gripper widths for humanoids.
    System-level invention; builds on UMI/HuMI but is a specific portable pipeline claimed as contribution 1.
  • Spatial Keypoint Retargeting (SKR)
    purpose: Map sparse human/predicted keypoints to robot root pose and joints via anisotropic leg scale and weighted IK.
    Explicit bridge module (§III.C); ablated vs GMR but not independently validated outside this pipeline.
  • Default 5-keypoint / augmented 7-keypoint whole-body task interface
    purpose: Compact action space separating task intent from joint control for hierarchical learning.
    Representation choice central to transfer claims; knee ablation supports utility but is paper-internal.

how reviews work

0 comments
Cite this review

Pith. "Pith review of HumanoidUMI: Bridging Robot-Free Demonstrations and Humanoid Whole-Body Manipulation." pith.science (2026). https://pith.science/paper/5YK7W5VP

@misc{pith2026260627239,
  author       = {Pith},
  title        = {Pith review of: HumanoidUMI: Bridging Robot-Free Demonstrations and Humanoid Whole-Body Manipulation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5YK7W5VP}},
  note         = {Machine review of arXiv:2606.27239}
}
read the original abstract

High-quality demonstration data are essential for humanoid robot skill learning, especially for whole-body behaviors that require coordinated perception, locomotion, and manipulation. Existing data-collection methods largely rely on robot teleoperation, which is constrained by hardware accessibility, operator expertise, and limited efficiency. Inspired by the Universal Manipulation Interface (UMI), we propose HumanoidUMI, a portable and robot-free framework for humanoid whole-body data collection. HumanoidUMI uses lightweight VR devices and UMI-inspired grippers to collect sparse human keypoint trajectories, wrist-view observations, and gripper actions. These demonstrations train a high-level policy to predict future keypoints, which are retargeted to robot-native whole-body references and executed by a whole-body controller. Experiments in five real-world scenarios demonstrate the effectiveness of the proposed framework and validate the collected demonstrations for transferable humanoid whole-body skill learning.

Figures

Figures reproduced from arXiv: 2606.27239 by the authors.

Figure 1
Figure 1. Overview of HumanoidUMI. HumanoidUMI provides a robot-free data collection and learning framework for humanoid whole-body skills. Human demonstrations are collected with portable VR–UMI devices and represented as sparse task-relevant spatial keypoints. A high-level policy predicts future keypoint motions and gripper actions, a Spatial Keypoint Retargeting module converts them into feasible humanoid whole-body refere… view at source ↗
Figure 2
Figure 2. HumanoidUMI Data Acquisition System. The data acquisition platform consists of a PICO-based motion capture setup, including two foot￾mounted trackers and one waist-mounted tracker, together with two instrumented grippers, each equipped with a fisheye camera. The system synchronously records multimodal observations, including wrist-view images from the fisheye cameras, human keypoint states obtained via the PICO SDK,… view at source ↗
Figure 3
Figure 3. HumanoidUMI Hierarchical Visuomotor Control. HumanoidUMI formulates humanoid visuomotor control as a three-stage hierarchy. A diffusion-based high-level policy predicts task-space keypoint trajectories and gripper commands from wrist-view images and lower-body proprioception. The spatial keypoint retargeting bridge maps these commands into a 36-dimensional robot-native motion representation, including the root pose … view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Conditional diffusion policy architecture. Wrist-view RGB images are encoded by DINOv2 and fused with lower-body proprioception and the diffusion step as the global condition. The diffusion model predicts action trajectories for TCPs and body-support keypoints. Action …
Figure 5
Figure 5. Figure 5: Spatial Keypoint Retargeting (SKR). SKR bridges high-level keypoint prediction and low-level whole-body control by converting five task-space keypoints, including the pelvis, two TCPs, and two feet, into robot-native whole-body references. Unlike global motion rescalin…
Figure 6
Figure 6. Figure 6: Real-world evaluation and ablation analysis of HumanoidUMI on three humanoid manipulation tasks. (a) Cluttered tabletop pick-and-place: the robot grasps bread and places it onto a target plate. (b) Bimanual vegetable collection: the robot uses the right hand to place a…
Figure 7
Figure 7. Figure 7: Real-world evaluation of the augmented seven-keypoint variant of HumanoidUMI on two humanoid whole-body manipulation tasks with a Unitree G1 robot. (a) Under-table waste disposal: the robot grasps a crumpled paper ball from the tabletop, steps backward, bends its knees…

Review history (2 revisions) →

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

25 extracted references · 4 linked inside Pith

  1. [1]

    Droid: A large-scale in-the-wild robot manipulation dataset,

    A. Khazatsky, K. Pertsch, S. Nair, A. Balakrishna, S. Dasari, S. Karam- cheti, S. Nasiriany, M. K. Srirama, L. Y . Chen, K. Ellis,et al., “Droid: A large-scale in-the-wild robot manipulation dataset,” inProceedings of Robotics: Science and Systems, Delft, Netherlands, July 2024

    2024

  2. [2]

    Twist2: Scalable, portable, and holistic humanoid data collection system,

    Y . Ze, S. Zhao, W. Wang, A. Kanazawa, R. Duan, P. Abbeel, G. Shi, J. Wu, and C. K. Liu, “Twist2: Scalable, portable, and holistic humanoid data collection system,” in2026 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2026

    2026

  3. [3]

    Humanplus: Humanoid shadowing and imitation from humans,

    Z. Fu, Q. Zhao, Q. Wu, G. Wetzstein, and C. Finn, “Humanplus: Humanoid shadowing and imitation from humans,” inProceedings of The 8th Conference on Robot Learning, ser. Proceedings of Machine Learning Research, vol. 270. PMLR, 2025, pp. 2828–2844

    2025

  4. [4]

    Clone: Closed-loop whole-body humanoid teleoperation for long-horizon tasks,

    Y . Li, Y . Lin, J. Cui, T. Liu, W. Liang, Y . Zhu, and S. Huang, “Clone: Closed-loop whole-body humanoid teleoperation for long-horizon tasks,” 2025. [Online]. Available: https://arxiv.org/abs/2506.08931

    arXiv 2025

  5. [5]

    Mobile-television: Predictive motion priors for humanoid whole-body control,

    C. Lu, X. Cheng, J. Li, S. Yang, M. Ji, C. Yuan, G. Yang, S. Yi, and X. Wang, “Mobile-television: Predictive motion priors for humanoid whole-body control,” in2025 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2025

    2025

  6. [6]

    Learning versatile humanoid manipulation with touch dreaming,

    Y . Niu, Z. Fang, B. Chen, S. Zhou, R. Senthilkumaran, H. Zhang, B. Chen, C. Qiu, H. E. Tseng, J. Francis, and D. Zhao, “Learning versatile humanoid manipulation with touch dreaming,” 2026. [Online]. Available: https://arxiv.org/abs/2604.13015

    arXiv 2026

  7. [7]

    Learning human-to-humanoid real-time whole-body teleoperation,

    T. He, Z. Luo, W. Xiao, C. Zhang, K. Kitani, C. Liu, and G. Shi, “Learning human-to-humanoid real-time whole-body teleoperation,” in2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2024, pp. 8944–8951

    2024

  8. [8]

    Omnih2o: Universal and dexterous human-to- humanoid whole-body teleoperation and learning,

    T. He, Z. Luo, X. He, W. Xiao, C. Zhang, W. Zhang, K. M. Kitani, C. Liu, and G. Shi, “Omnih2o: Universal and dexterous human-to- humanoid whole-body teleoperation and learning,” inProceedings of The 8th Conference on Robot Learning, ser. Proceedings of Machine Learning Research, vol. 270. PMLR, 2025, pp. 1516–1540

    2025

Show all 25 references
  1. [9]

    Universal manipulation interface: In-the-wild robot teaching without in-the-wild robots,

    C. Chi, Z. Xu, C. Pan, E. Cousineau, B. Burchfiel, S. Feng, R. Tedrake, and S. Song, “Universal manipulation interface: In-the-wild robot teaching without in-the-wild robots,” inProceedings of Robotics: Science and Systems (RSS), 2024

    2024

  2. [10]

    Omniumi: Towards physically grounded robot learning via human-aligned multimodal interaction,

    S. Luo, Y . Li, Y . Hu, C. Yu, C. Xu, J. Zhang, G. Yao, T. Huang, R. He, and Z. Wang, “Omniumi: Towards physically grounded robot learning via human-aligned multimodal interaction,”arXiv preprint arXiv:2604.10647, 2026

    2026 arXiv

  3. [11]

    Data scaling laws in imitation learning for robotic manipulation,

    F. Lin, Y . Hu, P. Sheng, C. Wen, J. You, and Y . Gao, “Data scaling laws in imitation learning for robotic manipulation,” inInternational Conference on Learning Representations (ICLR), 2025

    2025

  4. [12]

    In-the-wild compliant manipulation with umi-ft,

    H. Choi, Y . Hou, C. Pan, S. Hong, A. Patel, X. Xu, M. R. Cutkosky, and S. Song, “In-the-wild compliant manipulation with umi-ft,” in 2026 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2026

    2026

  5. [13]

    Activeumi: Robotic manipulation with active perception from robot-free human demonstrations,

    Q. Zeng, C. Li, J. St. John, Z. Zhou, J. Wen, G. Feng, Y . Zhu, and Y . Xu, “Activeumi: Robotic manipulation with active perception from robot-free human demonstrations,” 2025. [Online]. Available: https://arxiv.org/abs/2510.01607

    2025

  6. [14]

    Umi-on-legs: Making manipulation policies mobile with manipulation-centric whole-body controllers,

    H. Ha, Y . Gao, Z. Fu, J. Tan, and S. Song, “Umi-on-legs: Making manipulation policies mobile with manipulation-centric whole-body controllers,” inProceedings of The 8th Conference on Robot Learning, ser. Proceedings of Machine Learning Research, vol. 270. PMLR, 2025, pp. 5254–5270

    2025

  7. [15]

    Umi-on-air: Embodiment-aware guidance for embodiment-agnostic visuomotor policies,

    H. Gupta, X. Guo, H. Ha, C. Pan, M. Cao, D. Lee, S. Scherer, S. Song, and G. Shi, “Umi-on-air: Embodiment-aware guidance for embodiment-agnostic visuomotor policies,” in2026 IEEE Interna- tional Conference on Robotics and Automation (ICRA). IEEE, 2026

    2026

  8. [16]

    Hommi: Learning whole-body mobile manipulation from human demonstrations,

    X. Xu, J. Park, H. Zhang, E. Cousineau, A. Bhat, J. Barreiros, D. Wang, J. Bohg, and S. Song, “Hommi: Learning whole-body mobile manipulation from human demonstrations,” inProceedings of Robotics: Science and Systems (RSS), 2026

    2026

  9. [17]

    Humanoid manipulation interface: Humanoid whole-body manipulation from robot-free demonstrations,

    R. Nai, B. Zheng, J. Zhao, H. Zhu, S. Dai, Z. Chen, Y . Hu, Y . Hu, T. Zhang, C. Wen,et al., “Humanoid manipulation interface: Humanoid whole-body manipulation from robot-free demonstrations,” arXiv preprint arXiv:2602.06643, 2026

    2026

  10. [18]

    Learning fine-grained bimanual manipulation with low-cost hardware,

    T. Z. Zhao, V . Kumar, S. Levine, and C. Finn, “Learning fine-grained bimanual manipulation with low-cost hardware,” inProceedings of Robotics: Science and Systems, Daegu, Republic of Korea, July 2023

    2023

  11. [19]

    Mobile aloha: Learning bimanual mobile manipulation using low-cost whole-body teleoperation,

    Z. Fu, T. Z. Zhao, and C. Finn, “Mobile aloha: Learning bimanual mobile manipulation using low-cost whole-body teleoperation,” in Proceedings of The 8th Conference on Robot Learning, ser. Proceed- ings of Machine Learning Research, vol. 270. PMLR, 2025, pp. 4066–4083

    2025

  12. [20]

    Egohumanoid: Unlocking in-the-wild loco-manipulation with robot-free egocentric demonstration,

    M. Shi, S. Peng, J. Chen, H. Jiang, Y . Li, D. Huang, P. Luo, H. Li, and L. Chen, “Egohumanoid: Unlocking in-the-wild loco-manipulation with robot-free egocentric demonstration,” 2026. [Online]. Available: https://arxiv.org/abs/2602.10106

    2026 arXiv

  13. [21]

    Xrobotoolkit: A cross-platform framework for robot teleoperation,

    Z. Zhao, L. Yu, K. Jing, and N. Yang, “Xrobotoolkit: A cross-platform framework for robot teleoperation,” in2026 IEEE/SICE International Symposium on System Integration (SII). IEEE, 2026, pp. 15–20

    2026

  14. [22]

    Diffusion policy: Visuomotor policy learning via action diffusion,

    C. Chi, S. Feng, Y . Du, Z. Xu, E. Cousineau, B. Burchfiel, and S. Song, “Diffusion policy: Visuomotor policy learning via action diffusion,” in Proceedings of Robotics: Science and Systems (RSS), 2023

    2023

  15. [23]

    On the continuity of rotation representations in neural networks,

    Y . Zhou, C. Barnes, J. Lu, J. Yang, and H. Li, “On the continuity of rotation representations in neural networks,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 5745–5753

    2019

  16. [24]

    mjlab: A Lightweight Framework for GPU-Accelerated Robot Learn- ing,

    K. Zakka, Q. Liao, B. Yi, L. Le Lay, K. Sreenath, and P. Abbeel, “mjlab: A Lightweight Framework for GPU-Accelerated Robot Learn- ing,” 2026

    2026

  17. [25]

    Retargeting matters: General motion retargeting for humanoid motion tracking,

    J. P. Araujo, Y . Ze, P. Xu, J. Wu, and C. K. Liu, “Retargeting matters: General motion retargeting for humanoid motion tracking,” in2026 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2026

    2026

Pith tools

Reviewed July 12, 2026 · model on record in the stance chip above.