Pith. sign in

REVIEW 5 cited by

VideoFlow: A Conditional Flow-Based Model for Stochastic Video Generation

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1903.01434 v3 pith:REA56YYS submitted 2019-03-04 cs.CV cs.AIcs.LG

Manoj Kumar , Mohammad Babaeizadeh , Dumitru Erhan , Chelsea Finn , Sergey Levine , Laurent Dinh , Durk Kingma This is my paper · ORCID
classification cs.CVcs.AIcs.LG
keywords modelsvideogenerativeapproachdataeventsflow-basedfuture
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Generative models that can model and predict sequences of future events can, in principle, learn to capture complex real-world phenomena, such as physical interactions. However, a central challenge in video prediction is that the future is highly uncertain: a sequence of past observations of events can imply many possible futures. Although a number of recent works have studied probabilistic models that can represent uncertain futures, such models are either extremely expensive computationally as in the case of pixel-level autoregressive models, or do not directly optimize the likelihood of the data. To our knowledge, our work is the first to propose multi-frame video prediction with normalizing flows, which allows for direct optimization of the data likelihood, and produces high-quality stochastic predictions. We describe an approach for modeling the latent space dynamics, and demonstrate that flow-based generative models offer a viable and competitive approach to generative modelling of video.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Conditional Flow Variational Autoencoders for Structured Sequence Prediction

    cs.CV 2019-08 conditional novelty 6.0 of 10

    Conditional Flow VAEs replace the standard Gaussian prior of CVAEs with a conditional normalizing flow prior and two regularizations, improving multimodal sequence prediction.

  2. Likelihood Contribution based Multi-scale Architecture for Generative Flows

    cs.LG 2019-08 conditional novelty 6.0 of 10

    Using per-dimension log-likelihood contributions to decide which dimensions to factor out early improves bits/dim for RealNVP on CIFAR-10, ImageNet, and CelebA.

  3. Learning Variations in Human Motion via Mix-and-Match Perturbation

    cs.LG 2019-08 conditional novelty 6.0 of 10

    Mix-and-Match perturbation randomly replaces a subset of the RNN hidden state with noise, preventing conditional VAEs from ignoring the latent code and yielding more diverse human motion predictions.

  4. Video Interpolation and Prediction with Unsupervised Landmarks

    cs.CV 2019-09 conditional novelty 5.0 of 10

    Unsupervised 2D Gaussian landmarks, moved by a residual LSTM, enable long-range video interpolation and prediction while preserving foreground structure.

  5. Survey on Deep Neural Networks in Speech and Vision Systems

    cs.CV 2019-08 conditional

    A broad survey of deep learning architectures and systems for vision and speech, with an emphasis on mobile deployment and emerging applications, containing no new results.

Pith tools