Texture Synthesis with Recurrent Variational Auto-Encoder

We propose a recurrent variational auto-encoder for texture synthesis. A novel loss function, FLTBNK, is used for training the texture synthesizer. It is rotational and partially color invariant loss function. Unlike L2 loss, FLTBNK explicitly models the correlation of color intensity between pixels...

Full description

Saved in:
Bibliographic Details
Main Authors Chandra, Rohan, Grover, Sachin, Lee, Kyungjun, Meshry, Moustafa, Taha, Ahmed
Format Journal Article
LanguageEnglish
Published 23.12.2017
Subjects
Computer Science - Computer Vision and Pattern Recognition
Online AccessGet full text
DOI10.48550/arxiv.1712.08838

Cover

More Information
Summary:We propose a recurrent variational auto-encoder for texture synthesis. A novel loss function, FLTBNK, is used for training the texture synthesizer. It is rotational and partially color invariant loss function. Unlike L2 loss, FLTBNK explicitly models the correlation of color intensity between pixels. Our texture synthesizer generates neighboring tiles to expand a sample texture and is evaluated using various texture patterns from Describable Textures Dataset (DTD). We perform both quantitative and qualitative experiments with various loss functions to evaluate the performance of our proposed loss function (FLTBNK) --- a mini-human subject study is used for the qualitative evaluation.
DOI:10.48550/arxiv.1712.08838