Publication:
Synthesising Dynamic Textures using Convolutional Neural Networks

dc.contributor.authorFunke, Christina M.
dc.contributor.authorGatys, Leon A.
dc.contributor.authorEcker, Alexander S.
dc.contributor.authorBethge, Matthias
dc.date.accessioned2020-03-31T14:46:52Z
dc.date.available2020-03-31T14:46:52Z
dc.date.issued2017
dc.description.abstractHere we present a parametric model for dynamic textures. The model is based on spatiotemporal summary statistics computed from the feature representations of a Convolutional Neural Network (CNN) trained on object recognition. We demonstrate how the model can be used to synthesise new samples of dynamic textures and to predict motion in simple movies.
dc.format.extent9
dc.identifier.arxiv1702.07006v1
dc.identifier.urihttps://resolver.sub.uni-goettingen.de/purl?gro-2/63422
dc.language.isoen
dc.titleSynthesising Dynamic Textures using Convolutional Neural Networks
dc.typepreprint
dc.type.internalPublicationno
dspace.entity.typePublication

Files

Collections