Can I read Transposer: Universal Texture Synthesis Using Feature Maps as Transposed Convolution Filter on EtoBox?

Transposer: Universal Texture Synthesis Using Feature Maps as Transposed Convolution Filter by Liu, Guilin; Taori, Rohan; Wang, Ting-Chun; Yu, Zhiding; Liu, Shiqiu; Reda, Fitsum A.; Sapra, Karan; Tao, Andrew; Catanzaro, Bryan is a scholarly article available to read on EtoBox.

What is Transposer: Universal Texture Synthesis Using Feature Maps as Transposed Convolution Filter about?

Conventional CNNs for texture synthesis consist of a sequence of (de)-convolution and up/down-sampling layers, where each layer operates locally and lacks the ability to capture the long-term structural dependency required by texture synthesis. Thus, they often simply enlarge the input texture, rather than perform reasonable synthesis. As a compromise, many recent methods sacrifice generalizability by training and testing on the same single (or fixed set of) texture image(s), resulting in huge re-training time costs for unseen images. In this work, based on the discovery that the assembling/stitching operation in traditional texture synthesis is analogous to a transposed convolution operation, we propose a novel way of using transposed convolution operation. Specifically, we directly treat the whole encoded feature map of the input texture as transposed convolution filters and the features' self-similarity map, which captures the auto-correlation information, as input to the transposed convolution. Such a design allows our framework, once trained, to be generalizable to perform synthesis of unseen textures with a single forward pass in nearly real-time. Our method achieves state-of

Author

Liu, Guilin; Taori, Rohan; Wang, Ting-Chun; Yu, Zhiding; Liu, Shiqiu; Reda, Fitsum A.; Sapra, Karan; Tao, Andrew; Catanzaro, Bryan

Published

2020

Language

EN