MoReFlow: Motion Retargeting Learning through Unsupervised Flow Matching
By: Wontaek Kim, Tianyu Li, Sehoon Ha
Potential Business Impact:
Moves one character's dance to another.
Motion retargeting holds a premise of offering a larger set of motion data for characters and robots with different morphologies. Many prior works have approached this problem via either handcrafted constraints or paired motion datasets, limiting their applicability to humanoid characters or narrow behaviors such as locomotion. Moreover, they often assume a fixed notion of retargeting, overlooking domain-specific objectives like style preservation in animation or task-space alignment in robotics. In this work, we propose MoReFlow, Motion Retargeting via Flow Matching, an unsupervised framework that learns correspondences between characters' motion embedding spaces. Our method consists of two stages. First, we train tokenized motion embeddings for each character using a VQ-VAE, yielding compact latent representations. Then, we employ flow matching with conditional coupling to align the latent spaces across characters, which simultaneously learns conditioned and unconditioned matching to achieve robust but flexible retargeting. Once trained, MoReFlow enables flexible and reversible retargeting without requiring paired data. Experiments demonstrate that MoReFlow produces high-quality motions across diverse characters and tasks, offering improved controllability, generalization, and motion realism compared to the baselines.
Similar Papers
MoReFlow: Motion Retargeting Learning through Unsupervised Flow Matching
Graphics
Moves one character's dance to another.
H-MoRe: Learning Human-centric Motion Representation for Action Analysis
CV and Pattern Recognition
Teaches computers to understand human movement better.
FlowMo: Variance-Based Flow Guidance for Coherent Motion in Video Generation
CV and Pattern Recognition
Makes videos move more smoothly and realistically.