Synthesising Dynamic Textures using Convolutional Neural Networks
2017 | preprint
Jump to: Cite & Linked | Documents & Media | Details | Version history
Documents & Media
Details
- Authors
- Funke, Christina M.; Gatys, Leon A.; Ecker, Alexander S. ; Bethge, Matthias
- Abstract
- Here 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.
- Issue Date
- 2017
- Extent
- 9
- Language
- English