English

Human Motion Synthesis_ A Diffusion Approach for Motion Stitching and In-Betweening

Computer Vision and Pattern Recognition 2024-09-12 v1 Human-Computer Interaction Machine Learning

Abstract

Human motion generation is an important area of research in many fields. In this work, we tackle the problem of motion stitching and in-betweening. Current methods either require manual efforts, or are incapable of handling longer sequences. To address these challenges, we propose a diffusion model with a transformer-based denoiser to generate realistic human motion. Our method demonstrated strong performance in generating in-betweening sequences, transforming a variable number of input poses into smooth and realistic motion sequences consisting of 75 frames at 15 fps, resulting in a total duration of 5 seconds. We present the performance evaluation of our method using quantitative metrics such as Frechet Inception Distance (FID), Diversity, and Multimodality, along with visual assessments of the generated outputs.

Keywords

Cite

@article{arxiv.2409.06791,
  title  = {Human Motion Synthesis_ A Diffusion Approach for Motion Stitching and In-Betweening},
  author = {Michael Adewole and Oluwaseyi Giwa and Favour Nerrise and Martins Osifeko and Ajibola Oyedeji},
  journal= {arXiv preprint arXiv:2409.06791},
  year   = {2024}
}

Comments

12 pages, 5 figures, and 11 equations

R2 v1 2026-06-28T18:40:23.089Z