Skip to yearly menu bar Skip to main content


Generating Videos with Scene Dynamics

Carl Vondrick · Hamed Pirsiavash · Antonio Torralba

Area 5+6+7+8 #29

Keywords: [ Large Scale Learning and Big Data ] [ (Application) Computer Vision ] [ (Other) Unsupervised Learning Methods ]


We capitalize on large amounts of unlabeled video in order to learn a model of scene dynamics for both video recognition tasks (e.g. action classification) and video generation tasks (e.g. future prediction). We propose a generative adversarial network for video with a spatio-temporal convolutional architecture that untangles the scene's foreground from the background. Experiments suggest this model can generate tiny videos up to a second at full frame rate better than simple baselines, and we show its utility at predicting plausible futures of static images. Moreover, experiments and visualizations show the model internally learns useful features for recognizing actions with minimal supervision, suggesting scene dynamics are a promising signal for representation learning. We believe generative video models can impact many applications in video understanding and simulation.

Live content is unavailable. Log in and register to view live content