Goal-Driven Human Motion Synthesis in Diverse Task

Abstract

We propose a framework for goal-driven human motion generation, which can synthesize interaction-rich scenarios. Given target positions for key joints, our pipeline generates a natural full-body motion that approaches the goal in cluttered environments. Our pipeline solves the complex constraints in a tractable formulation by disentangling the process of motion generation into two stages. The first stage computes the trajectory of the key joints, such as hands and feet, to encourage the character to approach the target position while avoiding possible physical violation. We demonstrate that diffusion-based guidance sampling can flexibly adapt to the local scene context while satisfying the target-goal conditions. Then, the subsequent second stage can easily generate plausible full-body motion that traverses the key joint trajectories. The proposed pipeline applies to various scenarios that have to account for 3D scene geometry and body joint configurations concurrently.