Rigging and skinning are essential steps to create realistic 3D animations, often requiring significant expertise and manual effort. Traditional attempts at automating these processes rely heavily on geometric heuristics and often struggle with objects of complex geometry. Recent data-driven approaches show potential for better generality, but are often constrained by limited training data.
We present the Anymate Dataset, a large-scale dataset of 230K 3D assets paired with expert-crafted rigging and skinning information— 70 times larger than existing datasets. Using this dataset, we propose a learning-based auto-rigging framework with three sequential modules for joint, connectivity, and skinning weight prediction. We systematically de sign and experiment with various architectures as baselines for each module and conduct comprehensive evaluations on our dataset to compare their performance. Our models significantly outperform existing methods, providing a foundation for comparing future methods in automated rigging and skinning.
(Sample 1)
(Sample 5)
(Sample 9)
(Sample 2)
(Sample 6)
(Sample 10)
(Sample 3)
(Sample 7)
(Sample 11)
(Sample 4)
(Sample 8)
(Sample 12)
There's a lot of excellent work that was introduced around the same time as ours.
RigAnything, MagicArticulate and One Model to Rig Them All employ autoregressive model for skeleton prediction.
MagicArticulate introduces the Articulation-XL Dataset derived from Objaverse-XL and comprising over 33,000 models annotated with rigging details.
One Model to Rig Them All contributes two datasets: Rig-XL Dataset a curated subset of about 14,000 rigged models from Objaverse-XL and the VRoid Dataset, a collection of 2,000 stylized characters from VRoidHub
Together, these works highlight the growing emphasis on scalable, data-driven rigging solutions.