TokenMotion: Decoupled Motion Control via Token Disentanglement for Human-centric Video Generation
This work addresses the problem of limited motion control in video generation for creative production applications, representing a significant but incremental advancement.
The paper tackles the challenge of jointly controlling camera movements and human poses in human-centric video generation, such as for creative production, by introducing TokenMotion, a DiT-based framework that enables fine-grained control and outperforms state-of-the-art methods in these tasks.
Human-centric motion control in video generation remains a critical challenge, particularly when jointly controlling camera movements and human poses in scenarios like the iconic Grammy Glambot moment. While recent video diffusion models have made significant progress, existing approaches struggle with limited motion representations and inadequate integration of camera and human motion controls. In this work, we present TokenMotion, the first DiT-based video diffusion framework that enables fine-grained control over camera motion, human motion, and their joint interaction. We represent camera trajectories and human poses as spatio-temporal tokens to enable local control granularity. Our approach introduces a unified modeling framework utilizing a decouple-and-fuse strategy, bridged by a human-aware dynamic mask that effectively handles the spatially-and-temporally varying nature of combined motion signals. Through extensive experiments, we demonstrate TokenMotion's effectiveness across both text-to-video and image-to-video paradigms, consistently outperforming current state-of-the-art methods in human-centric motion control tasks. Our work represents a significant advancement in controllable video generation, with particular relevance for creative production applications.