Texture Attribute Synthesis and Transfer Using Feed-Forward CNNs
Authors:
- Thomas IrmerTIThomas Irmer
- Tobias GlasmachersTGTobias Glasmachers
- Subhransu MajiSMSubhransu Maji
- 1 March 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 852-861
Abstract
We present a novel technique for texture synthesis and style transfer based on convolutional neural networks (CNNs). Our method learns feed-forward image generators that correspond to specification of styles and textures in terms of high-level describable attributes such as 'striped', 'dotted', or 'veined'. Two key conceptual advantages over template-based approaches are that attributes can be analyzed and activated individually, while a template image necessarily represents a simultaneous specification of many attributes, and that attributes can combine aspects of many texture templates allowing flexibility in the generation process. Once the attribute-wise networks are trained, applications to texture synthesis and style transfer are fast, allowing for real-time video processing.Keywords
This publication has 16 references indexed in Scilit:
- Deep Filter Banks for Texture Recognition, Description, and SegmentationInternational Journal of Computer Vision, 2016
- Bilinear CNN Models for Fine-Grained Visual Recognition2015
- Learning to generate chairs with convolutional neural networks2015
- Caffe2014
- CNN Features Off-the-Shelf: An Astounding Baseline for Recognition2014
- Stochastic Gradient Descent Tricks2012
- A committee of neural networks for traffic sign classification2011
- Fast texture synthesis using tree-structured vector quantization2000
- Texture synthesis by non-parametric sampling1999
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998