One Model to Rig Them All: Diverse Skeleton Rigging with UniRig
Open accessThe DOI matches its Crossref record; 57 of the 57 references it cites were found in a registry or an index.
Abstract
The rapid evolution of 3D content creation, encompassing both AI-powered methods and traditional workflows, is driving an unprecedented demand for automated rigging solutions that can keep pace with the increasing complexity and diversity of 3D models. We introduce UniRig , a novel, unified framework for automatic skeletal rigging that leverages the power of large autoregressive models and a bone-point cross-attention mechanism to generate both high-quality skeletons and skinning weights. Unlike previous methods that struggle with complex or non-standard topologies, UniRig accurately predicts topologically valid skeleton structures thanks to a new Skeleton Tree Tokenization method that efficiently encodes hierarchical relationships within the skeleton. To train and evaluate UniRig, we present Rig-XL , a new large-scale dataset of over 14,000 rigged 3D models spanning a wide range of categories. UniRig significantly outperforms state-of-the-art academic and commercial methods, achieving a 215% improvement in rigging accuracy and a 194% improvement in motion accuracy on challenging datasets. Our method works seamlessly across diverse object categories, from detailed anime characters to complex organic and inorganic structures, demonstrating its versatility and robustness. By automating the tedious and time-consuming rigging process, UniRig has the potential to speed up animation pipelines with unprecedented ease and efficiency. Project Page: https://zjp-shadow.github.io/works/UniRig/
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The text we found for this paper looks like another document, so the self-check wasn’t run.
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
Decoupled Weight Decay Regularization
Loshchilov, I., & Hutter, F. (2017) · arXiv · doi:10.48550/arxiv.1711.05101 · Cited by 12,522
Representation and recognition of the spatial organization of three-dimensional shapes
Marr, D., & Nishihara, H. K. (1978) · Proceedings of the Royal Society of London. Series B. Biological Sciences · doi:10.1098/rspb.1978.0020 · Cited by 2,158
Rényi Divergence and Kullback-Leibler Divergence
van Erven, T., & Harremoes, P. (2014) · IEEE Transactions on Information Theory · doi:10.1109/tit.2014.2320500 · Cited by 1,555
Objaverse-XL: A Universe of 10M+ 3D Objects
Deitke, M., Liu, R., Wallingford, M., Ngo, H., Michel, O., Kusupati, A., Fan, A., Laforte, C., Voleti, V., Gadre, S. Y., et al. (2023) · neural information processing systems · doi:10.52202/075280-1554 · Cited by 848
SMPL: A Skinned Multi-Person Linear Model
Loper, M., Mahmood, N., Romero, J., Pons-Moll, G., & Black, M. J. (2023) · Seminal Graphics Papers: Pushing the Boundaries, Volume 2 · doi:10.1145/3596711.3596800 · Cited by 768
Surface Reconstruction by Voronoi Filtering
Amenta, N., & Bern, M. (1999) · Discrete & Computational Geometry · doi:10.1007/pl00009475 · Cited by 705
Point Transformer V3: Simpler, Faster, Stronger
Wu, X., Jiang, L., Wang, P.-S., Liu, Z., Liu, X., Qiao, Y., Ouyang, W., He, T., & Zhao, H. (2024) · 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52733.2024.00463 · Cited by 648 · cited 1×
“For instance, knowing that a bone belongs to a specific template (e.g., Mixamo) allows for efficient motion retargeting, as the mapping between the template and the target skeleton is already known.”
Phase-functioned neural networks for character control
Holden, D., Komura, T., & Saito, J. (2017) · ACM Transactions on Graphics · doi:10.1145/3072959.3073663 · Cited by 546 · cited 1×
“The rapid advancements in AI-driven 3D content creation [Holden et al. 2017; Peng et al. 2024; Poole et al. 2022; Siddiqui et al. 2024; Yu et al. 2024; Zhang et al. 2024b] are revolutionizing computer graphics, enabling the generation of complex 3D models at an unprecedented scale and speed.”
DreamFusion: Text-to-3D using 2D Diffusion
Poole, B., Jain, A., Barron, J. T., & Mildenhall, B. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2209.14988 · Cited by 474
Research works printed but not linked: 2
- [7] Blender. 2018. Blender - a 3D modelling and rendering package. Blender Foundation, Stichting Blender Foundation, Amsterdam. http://www.blender.org Google Scholar
- [56] Susan Zhang, Stephen Roller, Naman Goyal, Mikel Artetxe, Moya Chen, Shuohui Chen, Christopher Dewan, Mona Diab, Xian Li, Xi Victoria Lin, et al. 2022. Opt: Open pre-trained transformer language models. arXiv preprint arXiv:2205.01068 (2022). Google Scholar Cited By View all Chiba KMatsumoto HNakamura ANishide S(2026)Development of a User-Driven 3D Model Generation Support System and Motion Capture-Based Action ImitationICEIB 202610.3390/engproc2026159022(22)Online publication date: 21-Oct-2026 https://doi.org/10.3390/engproc2026159022 Zhang JPu CGuo MCao YHu S(2026)SkinTokens: A Learned Compact Representation for Unified Autoregressive RiggingACM Transactions on Graphics10.1145/3859874Online publication date: 21-Oct-2026 https://dl.acm.org/doi/10.1145/3859874 Wang AAo YWu STang C(2026)SK-Adapter: Skeleton-Based Structural Control for Native 3D GenerationComputer Vision – ECCV 202610.1007/978-3-032-37242-0_15(258-276)Online publication date: 30-Sep-2026 https://doi.org/10.1007/978-3-032-37242-0_15 Show More Cited By Index Terms Index Terms (Premium feature) One Model to Rig Them All: Diverse Skeleton Rigging with UniRig Computing methodologies Artificial intelligence Computer vision Computer vision representations Shape representations Computer graphics Animation Motion processing Recommendations Recommendations (Premium feature) Rig it Right! Maya Animation Rigging Concepts Read More Rig it Right! Maya Animation Rigging Concepts, 2nd edition Read More Rig it Right! Maya Animation Rigging Concepts, 2nd edition Read More Comments Comments (Premium feature) Information & Contributors Information Published In cover image ACM Transactions on Graphics ACM Transactions on Graphics Volume 44, Issue 4 August 2025 · arXiv · doi:10.3390/engproc2026159022(22)online
Cited by
Zhang, H., Xu, H., Feng, C., et al. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00623 · Cited by 3
“Among various techniques, Linear Blend Skinning (LBS) remains the most widely used due to its simplicity and computational efficiency [18].”
iTACO: Interactable Digital Twins of Articulated Objects from Casually Captured RGBD Videos
Peng, W., Lv, J., Lu, C., & Savva, M. (2026) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv69130.2026.00056 · Cited by 22
“In contrast, our method does not rely on object retrieval or fine-tuning, and is inherently more general and scalable.”
Comparing 3D Data Representations for Keypoint Estimation in Humanoid Characters
Lefèvre, J.-E., Cheynel, T., & Daniel, T. (2025) · SIGGRAPH Asia Posters · doi:10.1145/3757374.3771481 · Cited by 1
MorphGS: Morphology-Adaptive Articulated 3D Motion Transfer from Videos
Kim, T., Na, Y., Lee, J., Lee, S., Sung, M., & Yoon, S. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37314-4_12 · Cited by 4
Θεοδωρόπουλος, Α. (2026) · Virtual Worlds · doi:10.3390/virtualworlds5020020
“Autoregressive transformers, as implemented in UniRig, predict diverse skeleton rigging and skinning weights without manual modeling [34].”
Zhang, X. (2026) · International Conference on Inventive Computation Technologies (ICICT) · doi:10.1109/icict68280.2026.11511120
“The UniRig framework was proposed by Zhang et al. [8].”
ACT: A Unified Framework for Rigging and Animating Characters with Arbitrary Topologies
Long, P., Wang, W., Zhao, Q., Guo, X., Pan, X., Zhang, Q., Zhou, J., Hu, T., Yang, W., Xu, L., et al. (2026) · ACM Transactions on Graphics · doi:10.1145/3811392 · Cited by 1
“Current automatic rigging methods, ranging from graph-based approaches like RigNet [Xu et al. 2020] to recent transformer-based systems, i.e., UniRig [Zhang et al. 2025a] and MagicArticu-late [Song et al. 2025b], typically predict skeletal structure in isolation.”
InfluentialAniGen: Unified S 3 Fields for Animatable 3D Asset Generation
Huang, Y.-H., Zou, Z.-X., He, Y., Chang, C., Pu, C.-F., Yang, Z., Guo, Y.-C., Cao, Y.-P., & Qi, X. (2026) · ACM Transactions on Graphics · doi:10.1145/3811297
“…strategy to bridge this gap is a sequential “generate-then-rig” pipeline: first synthesize a static mesh using a state-of-the-art 3D generator [Xiang et al. 2025b], followed by an off-the-shelf auto-rigging algorithm [Deng et al. 2025; Liu et al. 2025; Song et al. 2025a; Zhang et al. 2025].”
Self-citationSkeletonGaussian: Editable 4D Generation Through Gaussian Skeletonization
Wu, L., Zhu, R., Ai, Y., & Zhang, T. (2026) · Computational Visual Media · doi:10.26599/cvm.2026.9450557 · Cited by 1
Building a cross-platform WebXR heritage application from a drone survey using a language modelPreprint
Islam, R., & Almutery, S. (2026) · Research Square · doi:10.21203/rs.3.rs-11062255/v1
Skin Tokens: A Learned Compact Representation for Unified Autoregressive Rigging
Zhang, J.-P., Pu, C.-F., Guo, M.-H., Cao, Y.-P., & Hu, S.-M. (2026) · arXiv · doi:10.48550/arxiv.2602.04805
“The discrete and compact nature of SkinTokens enables us to move beyond the limitations of decoupled, multi-stage pipelines [Deng et al. 2025; Guo et al. 2025a,b; Liu et al. 2025b; Song et al. 2025a,b; Zhang et al. 2025b].”
InfluentialSelf-citationViP-Rig: Visual-Prompted Controllable Rigging
Qin, Z., Sun, M., Mao, Y., Xu, J., Guo, J., Hu, C., Zhao, W., Jiang, J., & Liu, X. (2026) · arXiv · doi:10.48550/arxiv.2607.27982
“UniRig allows users to edit a generated 3D hierarchy before continuing generation (Zhang et al. 2025), but requires professional 3D manipulation.”
InfluentialRigel3D: Rig-aware Latents for Animation-Ready 3D Asset Generation
Chatzis, N., Loizou, M., & Kalogerakis, E. (2026) · arXiv · doi:10.48550/arxiv.2605.13129
Metadata
- Authors
- Zhang, J.-P., Pu, C.-F., Guo, M.-H., Cao, Y.-P., & Hu, S.-M.
- Affiliations
- Tsinghua University (CN)
- VAST, BEIJING, China
- Venue
- ACM Transactions on Graphics
- Published
- 2025-07-27
- Volume
- 44
- Issue
- 4
- Pages
- 1-18
- ISSN
- 0730-0301 (ISSN-L)
- 1557-7368
- OpenAlex
- W4412673761
- Cited by
- 12
- References
- 33
- Citations by year
- 2026: 10 · 2025: 2
- FWCI
- 5.22
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- ACM Transactions on Graphics (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University) (cc-by)
- Funders
- the Natural Science Foundation of China (62220106003, 623B2057)
- Topics
- Physical Sciences › Computer Science › Computer Graphics and Computer-Aided Design › 3D Shape Modeling and Analysis
- Physical Sciences › Computer Science › Computer Graphics and Computer-Aided Design › Human Motion and Animation
- Physical Sciences › Computer Science › Computer Graphics and Computer-Aided Design › Computer Graphics and Visualization Techniques
- Keywords
- skinning weights, cross-attention, 3D character animation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A second model labelled each claim on 2026-10-29 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.