Nicolas Gonthier, Yann Gousseau and Saïd Ladjal

Article - PDF - Code - Supplementary Materials - Synthesis Results (1024 * 1024) - Synthesis Results (2048 * 2048)

Abstract

The field of texture synthesis has witnessed important progresses over the last years, most notably through the use of Convolutional Neural Networks. However, neural synthesis methods still struggle to reproduce large scale structures, especially with high resolution textures. To address this issue, we first introduce a simple multi-resolution framework that efficiently accounts for long-range dependency. Then, we show that additional statistical constraints further improve the reproduction of textures with strong regularity. This can be achieved by constraining both the Gram matrices of a neural network and the power spectrum of the image. Alternatively one may constrain only the autocorrelation of the features of the network and drop the Gram matrices constraints. In an experimental part, the proposed methods are then extensively tested and compared to alternative approaches, both in an unsupervised way and through a user study. Experiments show the interest of the multi-scale scheme for high resolution textures and the interest of combining it with additional constraints for regular textures.

Keywords

• Texture Synthesis
• Deep learning
• Power Spectrum
• Perceptual test

Remark

The paper is in low resolution but the high resolution texture synthesis can be found here : Results in (1024 * 1024) - Results in (2048 * 2048) for a fair comparison.

Recommended citation: Gonthier, N., Gousseau Y. and Ladjal S. (2022). “High resolution neural texture synthesis with long range constraints” Journal of Mathematical Imaging and Vision volume 64.