Latent Video Transformer | Alexey Artemov

Abstract

The video generation task can be formulated as a prediction of future video frames given some past frames. Recent generative models for videos face the problem of high computational requirements. Some models require up to 512 Tensor Processing Units for parallel training. In this work, we address this problem via modeling the dynamics in a latent space. After the transformation of frames into the latent space, our model predicts latent representation for the next frames in an autoregressive manner. We demonstrate the performance of our approach on BAIR Robot Pushing and Kinetics-600 datasets. The approach tends to reduce requirements to 8 Graphical Processing Units for training the models while maintaining comparable generation quality.

Resources

Fulltext PDF: Latent Video Transformer
Source code: https://github.com/rakhimovv/lvt
Slides: Slides

Citing:

@inproceedings{2021-lvt,
  title={Latent video transformer},
  author={Rakhimov, Ruslan and Volkhonskiy, Denis and Artemov, Alexey and Zorin, Denis and Burnaev, Evgeny},
  booktitle={16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2021},
  pages={101--112},
  year={2021},
  organization={SciTePress}
}

Ruslan Rakhimov	Denis Volkhonskiy	Alexey Artemov	Denis Zorin
Evgeny Burnaev