Generative Motion Infilling From Imprecisely Timed Keyframes
This addresses the tedious and challenging task of specifying keyframe timings in motion synthesis for animation or robotics, representing an incremental improvement over existing learned motion-inbetweening methods.
The paper tackles the problem of generating high-quality motion from imprecisely timed keyframes, which degrade motion quality in existing methods, by introducing a model that retimes constraints and adds pose details, resulting in diverse, realistic motions with plausible timing.
Keyframes are a standard representation for kinematic motion specification. Recent learned motion-inbetweening methods use keyframes as a way to control generative motion models, and are trained to generate life-like motion that matches the exact poses and timings of input keyframes. However, the quality of generated motion may degrade if the timing of these constraints is not perfectly consistent with the desired motion. Unfortunately, correctly specifying keyframe timings is a tedious and challenging task in practice. Our goal is to create a system that synthesizes high-quality motion from keyframes, even if keyframes are imprecisely timed. We present a method that allows constraints to be retimed as part of the generation process. Specifically, we introduce a novel model architecture that explicitly outputs a time-warping function to correct mistimed keyframes, and spatial residuals that add pose details. We demonstrate how our method can automatically turn approximately timed keyframe constraints into diverse, realistic motions with plausible timing and detailed submovements.