Skip to Main content Skip to Navigation

Stochastic Latent Residual Video Prediction

Abstract : Designing video prediction models that account for the inherent uncertainty of the future is challenging. Most works in the literature are based on stochastic image-autoregressive recurrent networks, which raises several performance and applicability issues. An alternative is to use fully latent temporal models which untie frame synthesis and temporal dynamics. However, no such model for stochastic video prediction has been proposed in the literature yet, due to design and training difficulties. In this paper, we overcome these difficulties by introducing a novel stochastic temporal model whose dynamics are governed in a latent space by a residual update rule. This first-order scheme is motivated by discretization schemes of differential equations. It naturally models video dynamics as it allows our simpler, more interpretable, latent model to outperform prior state-of-the-art methods on challenging datasets.
Complete list of metadatas

Cited literature [11 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02484182
Contributor : Jean-Yves Franceschi <>
Submitted on : Wednesday, February 19, 2020 - 10:06:10 AM
Last modification on : Sunday, February 23, 2020 - 1:24:01 AM

Files

main.pdf
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

  • HAL Id : hal-02484182, version 1
  • ARXIV : 2002.09219

Citation

Jean-Yves Franceschi, Edouard Delasalles, Mickaël Chen, Sylvain Lamprier, Patrick Gallinari. Stochastic Latent Residual Video Prediction. 2020. ⟨hal-02484182⟩

Share

Metrics

Record views

85

Files downloads

48