Conditional motion in-betweening

Motion in-betweening (MIB) is a process of generating intermediate skeletal movement between the given start and target poses while preserving naturalness motion, such as periodic footstep motion walking. Although state-of-the-art MIB methods are capable producing plausible motions sparse key-poses, they often lack controllability to generate satisfying semantic contexts required in practical applications. We focus on method that can handle pose or conditioned tasks using unified model. also present augmentation improve quality pose-conditioned generation via defining distribution over smooth trajectories. Our proposed outperforms existing prediction errors providing additional controllability.