Conditional Motion In-betweening

Summary

Conditional Motion In-betweening generates natural skeletal motion between start and target poses while adding semantic control absent from earlier methods. A single model supports both pose-conditioned and semantic-conditioned in-betweening, with smooth-trajectory augmentation improving pose-conditioned generation. It outperforms prior state-of-the-art methods on pose prediction error while providing additional controllability. (Kim et al., 2022)

Resources

Project Page · HTML · PDF · arXiv · GitHub · Video

References

2022

Pattern Recogn.

Conditional Motion In-betweening

Jihoon Kim, Taehyun Byun, Seungyoun Shin, Jungdam Won, and Sungjoon Choi

Pattern Recognition, Dec 2022

DOI arXiv Bib HTML PDF Video Code Website

@article{kim2022conditional,
  title = {Conditional Motion In-betweening},
  author = {Kim, Jihoon and Byun, Taehyun and Shin, Seungyoun and Won, Jungdam and Choi, Sungjoon},
  journal = {Pattern Recognition},
  volume = {132},
  pages = {108894},
  month = dec,
  year = {2022},
  publisher = {Elsevier},
  doi = {10.1016/j.patcog.2022.108894},
}