pith. sign in

arxiv: 2606.19156 · v1 · pith:VVEEVWN3new · submitted 2026-06-17 · 💻 cs.CV

Hand-4DGS: Feed-Forward 3D Gaussian Splatting for 4D Hand Reconstruction from Egocentric Videos

Jeongmin Bae , Seoha Kim , Marc Pollefeys , Mahdi Rad , Youngjung Uh , Taein Kwon This is my paper

Pith reviewed 2026-06-26 21:09 UTC · model grok-4.3

classification 💻 cs.CV
keywords 4D hand reconstructionegocentric video3D Gaussian splattingfeed-forward networkdynamic reconstructionmesh priorsAR/VR hand tracking
0
0 comments X

The pith

A feed-forward network with mesh-guided Gaussian splatting reconstructs dynamic 4D hands from single egocentric videos at roughly 60 frames per second.

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

The paper introduces Hand-4DGS to produce time-varying 3D hand models directly from head-mounted camera footage. It tackles fast head motion, rapid finger movements, heavy occlusions, and single-view depth ambiguity by predicting 3D Gaussians in one forward pass rather than through per-sequence optimization. The system adds a hand mesh as structural guidance and temporal convolutions to capture motion across frames. Evaluation on the H2O and ARCTIC datasets shows improved results over prior baselines while using only 2D image supervision and avoiding the need for 3D pose labels. If the approach holds, it removes the requirement for slow test-time optimization or multi-camera rigs in everyday AR and VR settings.

Core claim

Hand-4DGS is the first feed-forward framework that reconstructs dynamic 4D hands directly from egocentric videos by combining a mesh-guided representation for structural priors with temporal convolutions to model motion, delivering inference at approximately 60 FPS and strong generalization without expensive 3D hand-pose ground truth.

What carries the argument

Mesh-guided 3D Gaussian representation predicted by a feed-forward network that also applies temporal convolutions across frames to encode dynamic hand motion.

If this is right

  • Reconstruction runs at interactive rates without per-video optimization.
  • Performance improves on two standard egocentric hand datasets relative to existing methods.
  • Training relies only on 2D image losses from Gaussian splatting, removing the need for 3D annotations.
  • The same architecture generalizes across different subjects and recording conditions in the tested datasets.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The feed-forward design could be combined with existing real-time mesh trackers to initialize the Gaussian prediction and further reduce error under heavy occlusion.
  • If the temporal convolution module scales, the same pattern might apply to reconstructing other fast-moving objects such as tools or faces from head-mounted video.
  • Deployment on consumer AR glasses becomes feasible once the network is quantized, because no test-time optimization or multi-view setup is required.

Load-bearing premise

Mesh guidance plus temporal convolutions will be enough to overcome single-view depth ambiguity and rapid motion in egocentric hand videos.

What would settle it

Quantitative comparison on a new egocentric test set containing faster head turns or more severe finger occlusions than H2O or ARCTIC, where the method produces higher reconstruction error than an optimized per-sequence Gaussian baseline.

Figures

Figures reproduced from arXiv: 2606.19156 by Jeongmin Bae, Mahdi Rad, Marc Pollefeys, Seoha Kim, Taein Kwon, Youngjung Uh.

Figure 1
Figure 1. Figure 1: Overview of Hand-4DGS. Our feed-forward framework reconstructs dy￾namic 4D hands from challenging single-view egocentric videos at ∼60 FPS, without the heavy optimization time typically required by Gaussian splatting methods. By in￾troducing a mesh-guided representation with vertex-anchored positional embeddings, Hand-4DGS effectively captures both detailed hand appearance and complex motion. Our approach … view at source ↗
Figure 2
Figure 2. Figure 2: Framework Overview. (a) To leverage structural priors from the hand mesh, we predict embeddings at each mesh vertex. These embeddings serve as conditional la￾tents for predicting Gaussian attributes such as color and scale. To capture higher res￾olution, we sample additional Gaussians across the mesh surface and interpolate their embeddings from neighboring vertices. (b) Per-token features from adjacent fr… view at source ↗
Figure 3
Figure 3. Figure 3: Reconstruction Fidelity on the H2O dataset. [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Reconstruction Fidelity on the ARCTIC dataset. [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Comparison on Novel Viewpoints. Despite training on single-view egocen￾tric videos, our method reconstructs accurate hand geometry from unseen viewpoints. Hand-4DGS (Per-Video) captures hand structure accurately, though appearance may show artifacts from training view occlusions. Hand-4DGS (Generalized) further im￾proves appearance through multi-sequence training. Baselines fail to maintain hand structure … view at source ↗
read the original abstract

Dynamic 3D hand reconstruction from egocentric videos is essential for next-generation computing platforms such as AR/VR and AI glasses. Despite its importance, most prior works focus either on multi-view 3D hand reconstruction or on 4D human body reconstruction. Egocentric 4D hand reconstruction remains challenging due to fast head motion, rapid hand dynamics, severe occlusions, and inherent ambiguity from single-view observations. To address these challenges, we introduce Hand-4DGS, the first feed-forward framework for reconstructing dynamic 4D hands directly from egocentric videos, enabling both fast (~60 FPS) inference and strong generalization. Our approach incorporates a mesh-guided representation for structural priors and temporal convolutions to model dynamic motion. We evaluate our framework on two challenging egocentric datasets, H2O and ARCTIC, and demonstrate significant improvements over baselines. Our method benefits from the generalization capability of feed-forward networks and effective 2D image supervision through Gaussian splatting, without requiring expensive 3D hand pose ground-truth annotations.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper introduces Hand-4DGS, the first feed-forward framework for reconstructing dynamic 4D hands directly from egocentric videos using 3D Gaussian Splatting. It incorporates a mesh-guided representation for structural priors and temporal convolutions to model dynamic motion. The method is evaluated on the H2O and ARCTIC datasets and claims significant improvements over baselines, fast inference at ~60 FPS, strong generalization, and the ability to train using only 2D image supervision without 3D hand pose ground-truth annotations. The approach targets challenges including fast head motion, rapid hand dynamics, severe occlusions, and single-view ambiguity.

Significance. If the quantitative results and generalization claims hold, the work would be significant for enabling real-time 4D hand reconstruction in egocentric AR/VR settings. The feed-forward design combined with Gaussian splatting for 2D supervision (avoiding expensive 3D annotations) and the explicit handling of temporal dynamics represent practical advances over prior multi-view or non-feed-forward methods. The reported ~60 FPS inference speed would be a notable engineering contribution if validated.

major comments (2)
  1. [Abstract] Abstract: The claim that the method is the 'first feed-forward framework' for egocentric 4D hand reconstruction requires explicit support via a comparison table or discussion in the related work section; without it, the novelty statement remains unsubstantiated.
  2. [Evaluation] Evaluation section (referenced in abstract): The abstract asserts 'significant improvements over baselines' on H2O and ARCTIC, but the provided text contains no quantitative metrics, ablation results, or baseline descriptions; this prevents assessment of whether the gains are load-bearing or merely incremental.
minor comments (2)
  1. Define all acronyms (e.g., 4DGS, FPS) on first use in the abstract and introduction.
  2. The abstract mentions 'mesh-guided representation' and 'temporal convolutions' but provides no equations or architectural diagram references; adding these in §3 would improve clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed review. We address the two major comments below and will revise the manuscript to strengthen the novelty claim and ensure quantitative results are clearly presented.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The claim that the method is the 'first feed-forward framework' for egocentric 4D hand reconstruction requires explicit support via a comparison table or discussion in the related work section; without it, the novelty statement remains unsubstantiated.

    Authors: We agree that the 'first' claim requires explicit substantiation to avoid unsubstantiated novelty statements. In the revised manuscript, we will expand the Related Work section with a dedicated paragraph and a comparison table. The table will categorize prior works by input type (egocentric vs. multi-view), output representation (implicit, explicit mesh, NeRF, 3DGS), and whether they are feed-forward or optimization-based, highlighting that no existing method combines feed-forward inference with 3D Gaussian Splatting for dynamic egocentric 4D hand reconstruction. revision: yes

  2. Referee: [Evaluation] Evaluation section (referenced in abstract): The abstract asserts 'significant improvements over baselines' on H2O and ARCTIC, but the provided text contains no quantitative metrics, ablation results, or baseline descriptions; this prevents assessment of whether the gains are load-bearing or merely incremental.

    Authors: The full manuscript contains a dedicated Evaluation section with quantitative metrics (including PSNR, SSIM, LPIPS, and hand-specific metrics), ablation studies on mesh guidance and temporal convolutions, and comparisons against baselines on both H2O and ARCTIC. To address the concern that these results are not sufficiently visible from the abstract, we will add a short summary of key quantitative gains (e.g., FPS and reconstruction quality deltas) directly into the abstract and ensure the Evaluation section is cross-referenced clearly. We maintain that the reported improvements are load-bearing given the feed-forward design and 2D-only supervision. revision: yes

Circularity Check

0 steps flagged

No circularity: method claims rest on architectural choices and empirical evaluation, not self-referential definitions or fitted inputs.

full rationale

The abstract and available description introduce Hand-4DGS as a feed-forward network that combines mesh-guided priors with temporal convolutions, trained and evaluated on H2O and ARCTIC datasets using 2D image supervision via Gaussian splatting. No equations, loss terms, or uniqueness theorems are shown that reduce a claimed prediction to a parameter fitted from the target quantity itself. No self-citation is invoked as the sole justification for a load-bearing premise. The derivation chain is therefore self-contained: the network architecture is proposed, trained end-to-end, and measured against baselines, without any step that is definitionally equivalent to its own input.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Only the abstract is available, so no specific free parameters, axioms, or invented entities can be identified from the text.

pith-pipeline@v0.9.1-grok · 5735 in / 1186 out tokens · 20045 ms · 2026-06-26T21:09:00.106221+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

57 extracted references · 12 canonical work pages

  1. [1]

    ICCV (2021)

    Cao, Z., Radosavovic, I., Kanazawa, A., Malik, J.: Reconstructing hand-object interactions in the wild. ICCV (2021)

    2021

  2. [2]

    In: CVPR (2024)

    Charatan, D., Li, S., Tagliasacchi, A., Sitzmann, V.: pixelsplat: 3d gaussian splats from image pairs for scalable generalizable 3d reconstruction. In: CVPR (2024)

    2024

  3. [3]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Chen, K., Mohan, S., Theiss, J., Oprea, S., Sridhar, S., Prakash, A.: Interactavatar: Modelinghand-faceinteractioninphotorealisticavatarswithdeformablegaussians. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 10410–10420 (2025)

    2025

  4. [4]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Chen, X., Wang, B., Shum, H.Y.: Hand avatar: Free-pose hand animation and rendering from monocular video. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 8683–8693 (2023)

    2023

  5. [5]

    arXiv preprint arXiv:2403.14627 (2024)

    Chen, Y., Xu, H., Zheng, C., Zhuang, B., Pollefeys, M., Geiger, A., Cham, T.J., Cai, J.: Mvsplat: Efficient 3d gaussian splatting from sparse multi-view images. arXiv preprint arXiv:2403.14627 (2024)

    work page arXiv 2024

  6. [6]

    Dong, Y., Liu, H., Wang, Q., Yang, J., Wang, W., Slabaugh, G., Yuan, S.: Hand- splat: Embedding-driven gaussian splatting for high-fidelity hand rendering (2025), https://arxiv.org/abs/2503.14736

    work page arXiv 2025

  7. [7]

    In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision

    Duran, E., Kocabas, M., Choutas, V., Fan, Z., Black, M.J.: Hmp: Hand motion priors for pose and shape estimation from video. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. pp. 6353–6363 (2024)

    2024

  8. [8]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Fan, Z., Parelli, M., Kadoglou, M.E., Kocabas, M., Chen, X., Black, M.J., Hilliges, O.: HOLD: Category-agnostic 3d reconstruction of interacting hands and objects from video. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 494–504 (2024)

    2024

  9. [9]

    In: Pro- ceedings IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2023)

    Fan, Z., Taheri, O., Tzionas, D., Kocabas, M., Kaufmann, M., Black, M.J., Hilliges, O.: ARCTIC: A dataset for dexterous bimanual hand-object manipulation. In: Pro- ceedings IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2023)

    2023

  10. [10]

    Eu- ropean Conference on Computer Vision (2024)

    Fang, Y.K.B., Dong, Y.L.H., Carrasco, C.Z.F.V., Xu, A.M.M.J., Kim, S.T.D., la Torre, A.P.F.D.: Generalizable human gaussians for sparse view synthesis. Eu- ropean Conference on Computer Vision (2024)

    2024

  11. [11]

    In: International Confer- ence on Computer Vision (ICCV) (2023)

    Goel, S., Pavlakos, G., Rajasegaran, J., Kanazawa*, A., Malik*, J.: Humans in 4D: Reconstructing and tracking humans with transformers. In: International Confer- ence on Computer Vision (ICCV) (2023)

    2023

  12. [12]

    In: CVPR (2023)

    Guo, Z., Zhou, W., Wang, M., Li, L., Li, H.: HandNeRF: Neural radiance fields for animatable interacting hands. In: CVPR (2023)

    2023

  13. [13]

    In: Proceed- ings of the IEEE/CVF conference on computer vision and pattern recognition

    Hasson, Y., Varol, G., Tzionas, D., Kalevatykh, I., Black, M.J., Laptev, I., Schmid, C.: Learning joint reconstruction of hands and manipulated objects. In: Proceed- ings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 11807–11816 (2019) 16 J. Bae et al

    2019

  14. [14]

    In: NeurIPS (2024)

    Hu,H.,Fan,Z.,Wu,T.,Xi,Y.,Lee,S.,Pavlakos,G.,Wang,Z.:Expressivegaussian human avatars from monocular rgb video. In: NeurIPS (2024)

    2024

  15. [15]

    Hu, Y., He, Y., Chen, J., Yuan, W., Qiu, K., Lin, Z., Zhu, S., Dong, Z., Zhang, J.: Forge4d: Feed-forward 4d human reconstruction and interpolation from uncal- ibrated sparse-view videos (2025),https://arxiv.org/abs/2509.24209

    work page arXiv 2025

  16. [16]

    arXiv preprint arXiv:2410.01425 (2024)

    Hu, Y., Liu, Z., Shao, J., Lin, Z., Zhang, J.: Eva-gaussian: 3d gaussian-based real- time human novel view synthesis under diverse camera settings. arXiv preprint arXiv:2410.01425 (2024)

    work page arXiv 2024

  17. [17]

    Advances in Neural Information Processing Systems37, 14127–14147 (2024)

    Huang, X., Li, H., Liu, W., Liang, X., Yan, Y., Cheng, Y., Gao, C.: Learning interaction-aware 3d gaussian splatting for one-shot hand avatars. Advances in Neural Information Processing Systems37, 14127–14147 (2024)

    2024

  18. [18]

    CVPR (2022)

    Jain, A., Mildenhall, B., Barron, J.T., Abbeel, P., Poole, B.: Zero-shot text-guided object generation with dream fields. CVPR (2022)

    2022

  19. [19]

    Pattern Recognition Letters 183, 155–164 (2024)

    Jiao, Z., Wang, X., Li, J., Gao, R., He, M., Liang, J., Xia, Z., Gao, Q.: Handformer: Hand pose reconstructing from a single rgb image. Pattern Recognition Letters 183, 155–164 (2024)

    2024

  20. [20]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Karunratanakul, K., Prokudin, S., Hilliges, O., Tang, S.: Harp: Personalized hand reconstruction from a monocular rgb video. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 12802–12813 (2023)

    2023

  21. [21]

    ACM Transactions on Graphics42(4) (July 2023),https://repo-sam.inria.fr/fungraph/3d-gaussian-splatting/

    Kerbl, B., Kopanas, G., Leimkühler, T., Drettakis, G.: 3d gaussian splatting for real-time radiance field rendering. ACM Transactions on Graphics42(4) (July 2023),https://repo-sam.inria.fr/fungraph/3d-gaussian-splatting/

    2023

  22. [22]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Kocabas, M., Chang, J.H.R., Gabriel, J., Tuzel, O., Ranjan, A.: Hugs: Human gaussian splats. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 505–515 (2024)

    2024

  23. [23]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)

    Kwon, T., Tekin, B., Stühmer, J., Bogo, F., Pollefeys, M.: H2o: Two hands ma- nipulating objects for first person interaction recognition. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). pp. 10138– 10148 (October 2021)

    2021

  24. [24]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Li, M., An, L., Zhang, H., Wu, L., Chen, F., Yu, T., Liu, Y.: Interacting attention graph for single image two-hand reconstruction. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 2761–2770 (2022)

    2022

  25. [25]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Liu, S., Jiang, H., Xu, J., Liu, S., Wang, X.: Semi-supervised 3d hand-object poses estimation with interactions in time. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 14687–14697 (2021)

    2021

  26. [26]

    In: CVPR (2023)

    Moon, G.: Bringing inputs to shared domains for 3D interacting hands recovery in the wild. In: CVPR (2023)

    2023

  27. [27]

    Moon, G., Xu, W., Joshi, R., Wu, C., Shiratori, T.: Authentic hand avatar from a phonescanviauniversalhandmodel.In:ProceedingsoftheIEEE/CVFConference on Computer Vision and Pattern Recognition. pp. 2029–2038 (2024)

    2029

  28. [28]

    Mundra, A., B R, M., Wang, J., Habermann, M., Theobalt, C., Elgharib, M.: Livehand: Real-time and photorealistic neural hand rendering (October 2023)

    2023

  29. [29]

    On, J., Gwak, K., Kang, G., Cha, J., Hwang, S., Hwang, H., Baek, S.: Bigs: Bi- manual category-agnostic interaction reconstruction from monocular videos via 3d gaussian splatting (2025),https://arxiv.org/abs/2504.09097

    work page arXiv 2025

  30. [30]

    In: Advances in Neural Information Processing Systems (NeurIPS) (2024)

    Pan, P., Su, Z., Lin, C., Fan, Z., Zhang, Y., Li, Z., Shen, T., Mu, Y., Liu, Y.: Humansplat: Generalizable single-image human gaussian splatting with structure priors. In: Advances in Neural Information Processing Systems (NeurIPS) (2024)

    2024

  31. [31]

    In: CVPR (2024) Hand-4DGS 17

    Pavlakos, G., Shan, D., Radosavovic, I., Kanazawa, A., Fouhey, D., Malik, J.: Reconstructing hands in 3D with transformers. In: CVPR (2024) Hand-4DGS 17

    2024

  32. [32]

    Pokhariya, C., Shah, I.N., Xing, A., Li, Z., Chen, K., Sharma, A., Sridhar, S.: Manus: Markerless hand-object grasp capture using articulated 3d gaussians (2023)

    2023

  33. [33]

    In: Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Potamias, R.A., Ploumpis, S., Moschoglou, S., Triantafyllou, V., Zafeiriou, S.: Handy: Towards a high fidelity 3d hand shape and appearance model. In: Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 4670–4680 (2023)

    2023

  34. [34]

    In: Proceedings of the Computer Vision and Pattern Recognition Conference

    Potamias, R.A., Zhang, J., Deng, J., Zafeiriou, S.: Wilor: End-to-end 3d hand localization and reconstruction in-the-wild. In: Proceedings of the Computer Vision and Pattern Recognition Conference. pp. 12242–12254 (2025)

    2025

  35. [35]

    In: European Conference on Computer Vision

    Qian, N., Wang, J., Mueller, F., Bernard, F., Golyanik, V., Theobalt, C.: Html: A parametric hand texture model for 3d hand reconstruction and personalization. In: European Conference on Computer Vision. pp. 54–71. Springer (2020)

    2020

  36. [36]

    In: Proceedings of Neural Information Processing Systems(NeurIPS) (Dec 2024)

    Ren, J., Xie, K., Mirzaei, A., Liang, H., Zeng, X., Kreis, K., Liu, Z., Torralba, A., Fidler, S., Kim, S.W., Ling, H.: L4gm: Large 4d gaussian reconstruction model. In: Proceedings of Neural Information Processing Systems(NeurIPS) (Dec 2024)

    2024

  37. [37]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Ren, P., Wen, C., Zheng, X., Xue, Z., Sun, H., Qi, Q., Wang, J., Liao, J.: Decoupled iterative refinement framework for interacting hands reconstruction from a single rgb image. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 8014–8025 (2023)

    2023

  38. [38]

    ACM Transactions on Graphics, (Proc

    Romero, J., Tzionas, D., Black, M.J.: Embodied hands: Modeling and capturing hands and bodies together. ACM Transactions on Graphics, (Proc. SIGGRAPH Asia)36(6) (Nov 2017)

    2017

  39. [39]

    In: IEEE International Conference on Computer Vision Workshops (2021)

    Rong, Y., Shiratori, T., Joo, H.: Frankmocap: A monocular 3d whole-body pose es- timation system via regression and integration. In: IEEE International Conference on Computer Vision Workshops (2021)

    2021

  40. [40]

    arXiv preprint arXiv:2106.05954 (2021)

    Spurr, A., Molchanov, P., Iqbal, U., Kautz, J., Hilliges, O.: Adversarial mo- tion modelling helps semi-supervised hand pose estimation. arXiv preprint arXiv:2106.05954 (2021)

    work page arXiv 2021

  41. [41]

    Sun, Z., Shen, X., Hu, Y., Zhong, Y., Zhou, X.: Jghand: Joint-driven animatable handavatervia3dgaussiansplatting(2025),https://arxiv.org/abs/2501.19088

    work page arXiv 2025

  42. [42]

    arxiv (2024)

    Szymanowicz, S., Insafutdinov, E., Zheng, C., Campbell, D., Henriques, J., Rup- precht,C.,Vedaldi,A.:Flash3d:Feed-forwardgeneralisable3dscenereconstruction from a single image. arxiv (2024)

    2024

  43. [43]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Wang, X., Kwon, T., Rad, M., Pan, B., Chakraborty, I., Andrist, S., Bohus, D., Feniello, A., Tekin, B., Frujeri, F.V., et al.: Holoassist: an egocentric human in- teraction dataset for interactive ai assistants in the real world. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 20270–20281 (2023)

    2023

  44. [44]

    In: ICLR (2025)

    Wen, J., Schwing, A., Wang, S.: LIFe-GoM: Generalizable Human Rendering with Learned Iterative Feedback Over Multi-Resolution Gaussians-on-Mesh. In: ICLR (2025)

    2025

  45. [45]

    In: CVPR (2025)

    Xiao, J., Zhang, Q., Nie, Y., Zhu, L., Zheng, W.S.: RoGSplat: Learning robust generalizable human gaussian splatting from sparse multi-view images. In: CVPR (2025)

    2025

  46. [46]

    Xu, Z., Li, Z., Dong, Z., Zhou, X., Newcombe, R., Lv, Z.: 4dgt: Learning a 4d gaussian transformer using real-world monocular videos (2025)

    2025

  47. [47]

    In: European Conference on Computer Vision

    Yang, J., Chang, H.J., Lee, S., Kwak, N.: Seqhand: Rgb-sequence-based 3d hand pose and shape estimation. In: European Conference on Computer Vision. pp. 122–139. Springer (2020)

    2020

  48. [48]

    arXiv preprint arXiv:2501.08329 (2025) 18 J

    Ye, Y., Feng, Y., Tehari, O., Feng, H., Tulsiani, S., Black, M.J.: Predicting 4d hand trajectory from monocular videos. arXiv preprint arXiv:2501.08329 (2025) 18 J. Bae et al

    work page arXiv 2025

  49. [49]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Yu, Z., Huang, S., Fang, C., Breckon, T.P., Wang, J.: Acr: Attention collaboration- based regressor for arbitrary two-hand reconstruction. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 12955– 12964 (2023)

    2023

  50. [50]

    In: arXiv preprint arXiv:2412.12861 (December 2024)

    Yu, Z., Zafeiriou, S., Birdal, T.: Dyn-hamr: Recovering 4d interacting hand motion from a dynamic camera. In: arXiv preprint arXiv:2412.12861 (December 2024)

    work page arXiv 2024

  51. [51]

    arXiv preprint arXiv:2505.03351 (2025)

    Zhang, D., Liu, Y., Lin, L., Zhu, Y., Li, Y., Qin, M., Li, Y., Wang, H.: Guava: Generalizable upper body 3d gaussian avatar. arXiv preprint arXiv:2505.03351 (2025)

    work page arXiv 2025

  52. [52]

    arXiv preprint arXiv:2501.02973 (2025)

    Zhang, J., Deng, J., Ma, C., Potamias, R.A.: Hawor: World-space hand motion reconstruction from egocentric videos. arXiv preprint arXiv:2501.02973 (2025)

    work page arXiv 2025

  53. [53]

    IEEE Transactions on Visualization & Computer Graphics pp

    Zhao, L., Lu, X., Fan, R., Im, S.K., Wang, L.: Gaussianhand: Real-time 3d gaus- sian rendering for hand avatar animation. IEEE Transactions on Visualization & Computer Graphics pp. 1–13 (Dec 5555)

  54. [54]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2024)

    Zheng, S., Zhou, B., Shao, R., Liu, B., Zhang, S., Nie, L., Liu, Y.: Gps-gaussian: Generalizable pixel-wise 3d gaussian splatting for real-time human novel view syn- thesis. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2024)

    2024

  55. [55]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Zheng, X., Wen, C., Su, Z., Xu, Z., Li, Z., Zhao, Y., Xue, Z.: Ohta: One-shot hand avatar via data-driven implicit priors. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 799–810 (2024)

    2024

  56. [56]

    arXiv preprint arXiv:2412.14963 (2024),https://arxiv.org/abs/2412.14963

    Zhuang, Y., Lv, J., Wen, H., Shuai, Q., Zeng, A., Zhu, H., Chen, S., Yang, Y., Cao, X., Liu, W.: Idol: Instant photorealistic 3d human creation from a single image. arXiv preprint arXiv:2412.14963 (2024),https://arxiv.org/abs/2412.14963

    work page arXiv 2024

  57. [57]

    In: 2022 International Conference on 3D Vision (3DV)

    Ziani, A., Fan, Z., Kocabas, M., Christen, S., Hilliges, O.: Tempclr: Reconstructing hands via time-coherent contrastive learning. In: 2022 International Conference on 3D Vision (3DV). pp. 627–636. IEEE (2022) Hand-4DGS 1 In the Appendix, we present in-the-wild video results (Section A). To further show the robustness and applicability of our framework, w...

    2022