AnyTop can generate motion for unseen skeletons with just three examples per topology
rsasaki0109 · x · 2026-07-23
AnyTop: motion diffusion for arbitrary character topologies
At SIGGRAPH 2025, the authors introduce AnyTop, a diffusion model for generating motion on characters with diverse skeletons using only skeletal structure as input.
What it does
- Uses a transformer-based denoising network designed for arbitrary skeleton learning.
- Injects topology information into attention.
- Adds textual joint descriptions to latent features so the model can learn semantic correspondences across skeletons.
Reported results
- Generalizes well with as few as 3 training examples per topology.
- Can produce motion for unseen skeletons.
- Learns a latent space useful for downstream tasks like:
- joint correspondence
- temporal segmentation
- motion editing
More from Research
- Kimi K3 finds 32 of 86 vulnerabilities for just $19.37 in benchmark run — zeeg · 2026-07-23
- A chart comparing AI detectors says most are barely better than a coin toss — burkov · 2026-07-23
- CoLT teaches multimodal models to think in latent steps and cuts inference 10.1x — jiqizhixin · 2026-07-23
- AI agent finds a refutation to a 30-year-old graph theory conjecture in Slack — veggie_eric · 2026-07-23
- Duke study says better memory, not a bigger model, lifted ARC-AGI-3 scores by 18 points — imjustnewatai · 2026-07-23
- Trace builds a reproducible RLVR environment for 11 visual reasoning domains — Md Tanvirul Alam · 2026-07-23