Get Started
Home
Topics
Search
Library
Research questionHow can motion-style transfer learn new styles from few paired examples without artifacts on unseen motions?Creating diverse virtual-character motion requires adapting styles to new movements, but limited paired data can make new styles difficult to learn and prone to artifacts.
AI
Machine Learning
Research Paper
Technology
Latest papersRecent research connected to this question, newest first.STyMo: Fast and Controllable Few-Shot Motion Style TransferThe source describes STyMo, which learns from seconds of paired motion data and trains in one to two minutes. It separates static posture from temporal dynamics, supports runtime control and per-body-region styling, and uses a stylizability gate for out-of-distribution motions. Evidence covers subtle emotional variations and exaggerated character archetypes, along with a released processed paired dataset.research paper · Sep 3, 2026
Related questions
How can few-shot authorship style transfer match an arbitrary author’s context-dependent style while preserving meaning?How can mask-free video virtual try-on maintain garment consistency under motion, occlusion, and changing viewpoints?How can human motion be retargeted across different humanoid morphologies without hand-crafted mappings or losing pose detail?How can a transferable image-style representation separate overall identity from fine-grained visual attributes?