Corpus ID: 2940008

Spatio-temporal video autoencoder with differentiable memory

@article{Patraucean2015SpatiotemporalVA,
  title={Spatio-temporal video autoencoder with differentiable memory},
  author={Viorica Patraucean and Ankur Handa and Roberto Cipolla},
  journal={ArXiv},
  year={2015},
  volume={abs/1511.06309}
}
  • Viorica Patraucean, Ankur Handa, Roberto Cipolla
  • Published in ArXiv 2015
  • Computer Science
  • We describe a new spatio-temporal video autoencoder, based on a classic spatial image autoencoder and a novel nested temporal autoencoder. The temporal encoder is represented by a differentiable visual memory composed of convolutional long short-term memory (LSTM) cells that integrate changes over time. Here we target motion changes and use as temporal decoder a robust optical flow prediction module together with an image sampler serving as built-in feedback loop. The architecture is end-to-end… CONTINUE READING

    Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 169 CITATIONS

    Video Pixel Networks

    VIEW 5 EXCERPTS
    CITES RESULTS & BACKGROUND
    HIGHLY INFLUENCED

    Recurrent Flow-Guided Semantic Forecasting

    VIEW 4 EXCERPTS
    CITES METHODS
    HIGHLY INFLUENCED

    Deep generative video prediction

    VIEW 11 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Deep learning with synthetic, temporal, and adversarial supervision

    VIEW 21 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Flexible Spatio-Temporal Networks for Video Prediction

    VIEW 10 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Anomaly Detection Using Predictive Convolutional Long Short-Term Memory Units

    VIEW 7 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    gvnn: Neural Network Library for Geometric Computer Vision

    VIEW 6 EXCERPTS
    CITES BACKGROUND & METHODS

    Geometry and Uncertainty in Deep Learning for Computer Vision

    VIEW 6 EXCERPTS
    CITES BACKGROUND
    HIGHLY INFLUENCED

    Spatial-Temporal Convolutional LSTMs for Tumor Growth Prediction by Learning 4D Longitudinal Patient Data

    VIEW 4 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    FILTER CITATIONS BY YEAR

    2016
    2020

    CITATION STATISTICS

    • 15 Highly Influenced Citations

    • Averaged 47 Citations per year from 2017 through 2019

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 42 REFERENCES

    HMDB: A large video database for human motion recognition

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    FlowNet: Learning Optical Flow with Convolutional Networks

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Learning to See by Moving

    VIEW 3 EXCERPTS
    HIGHLY INFLUENTIAL

    Unsupervised Learning of Video Representations using LSTMs

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Generating Sequences With Recurrent Neural Networks

    VIEW 8 EXCERPTS
    HIGHLY INFLUENTIAL

    A Database and Evaluation Methodology for Optical Flow

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Multi-dimensional Recurrent Neural Networks

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Neural basis for a powerful static motion illusion.

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Neuromorphic Microchips.

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL