A Sliced Wasserstein Loss for Neural Texture Synthesis
This addresses a specific bottleneck in texture synthesis for computer vision applications, offering an incremental improvement over existing methods.
The authors tackled the problem of measuring distribution distance in neural texture synthesis by proposing the Sliced Wasserstein Distance as a replacement for the Gram-matrix loss, achieving visually superior results.
We address the problem of computing a textural loss based on the statistics extracted from the feature activations of a convolutional neural network optimized for object recognition (e.g. VGG-19). The underlying mathematical problem is the measure of the distance between two distributions in feature space. The Gram-matrix loss is the ubiquitous approximation for this problem but it is subject to several shortcomings. Our goal is to promote the Sliced Wasserstein Distance as a replacement for it. It is theoretically proven,practical, simple to implement, and achieves results that are visually superior for texture synthesis by optimization or training generative neural networks.