Researchers have developed a novel two-stage framework to improve the synthesis of realistic reactive motions, particularly for duet dancing. The first stage utilizes a motion VQ-VAE with specialized encoders and a joint decoder to enhance representation and prevent motion inconsistencies. The second stage employs a contact-aware diffusion model that generates both motion and a contact matrix between individuals, enabling explicit modeling of their interactions. AI
Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →
IMPACT This research could lead to more realistic and interactive virtual characters in applications like gaming and animation.
RANK_REASON This is a research paper detailing a new framework for motion synthesis.