SILK: Smooth InterpoLation frameworK for motion in-betweening A Simplified Computational Approach

TL;DR

This paper presents SILK, a simple Transformer-based framework for motion in-betweening that emphasizes data modeling choices over model complexity, achieving realistic animations with a single encoder.

Contribution

Introducing a streamlined Transformer-based approach for motion in-betweening that highlights the importance of data choices over complex models.

Findings

Increasing data volume improves motion transition quality.

Pose representation significantly impacts results.

Velocity features enhance animation realism.

Abstract

Motion in-betweening is a crucial tool for animators, enabling intricate control over pose-level details in each keyframe. Recent machine learning solutions for motion in-betweening rely on complex models, incorporating skeleton-aware architectures or requiring multiple modules and training steps. In this work, we introduce a simple yet effective Transformer-based framework, employing a single Transformer encoder to synthesize realistic motions for motion in-betweening tasks. We find that data modeling choices play a significant role in improving in-betweening performance. Among others, we show that increasing data volume can yield equivalent or improved motion transitions, that the choice of pose representation is vital for achieving high-quality results, and that incorporating velocity input features enhances animation performance. These findings challenge the assumption that model complexity is the primary determinant of animation quality and provide insights into a more data-centric approach to motion interpolation. Additional videos and supplementary material are available at https://silk-paper.github.io.