Paper page - UniMate: One Unified Model to Animate Diverse Skeletons

Source: https://huggingface.co/papers/2609.05415

Abstract

UniMate is a unified diffusion transformer that generates articulated motion for arbitrary skeletons from text and rigged 3D assets without per-skeleton retraining, using topology-aware attention and a large curated motion dataset.

Recent advances in automatic rigging now deliver animation-ready 3D assets at scale, yet generating the motion to drive them remains a bottleneck. Existing learned animators are topology-constrained: they rely on category-specific templates or require per-skeleton fine-tuning and reference motions at inference. We present UniMate, a unified foundation model that synthesizes articulated motion for arbitrary skeletons from a rigged 3D asset and a text prompt, with no test-time optimization or per-skeleton retraining. UniMate introduces atopology-aware diffusion transformer, which integrates skeletal topology into attention via three mechanisms: (1) agraph-aware attention biasfrom pairwise joint relations and geodesic distances; (2) aspectral rotary position embeddinggeneralizing RoPE to arbitrary kinematic trees via thegraph Laplacian; and (3) aglobal topological conditionerattention-pooled from the rest-pose skeleton. We also curateUniML3D, 13,006 motion sequences spanning bipedal, quadrupedal, avian, marine, insectoid, serpentine, and articulated rigid objects with unified canonicalization and text pairing. Trained on this dataset, UniMate outperforms state-of-the-art baselines in quality, generalization, and efficiency, and supportszero-shot cross-topology transfer,in-betweening, expansion, andtext-guided editing. Our project page is available at https://linzhanmou.com/unimate/.

View arXiv pageView PDFProject pageGitHub49Add to collection

Get this paper in your agent:

hf papers read 2609\.05415

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2609.05415 in a model README.md to link it from this page.

Datasets citing this paper4

#### Linzhan/Truebones-ZOO-Annotations Viewer• Updated31 minutes ago • 3.62k • 1.36k #### Linzhan/Mixamo-Animations-Characters Viewer• Updated31 minutes ago • 2.56k • 1.22k • 1 #### Linzhan/Objaverse-XL-Rigged-Animated Viewer• Updated31 minutes ago • 23.6k • 186 #### Linzhan/Objaverse-XL-Rigged-Animated-Renders Viewer• Updated31 minutes ago • 4 • 63

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2609.05415 in a Space README.md to link it from this page.

Collections including this paper1