-
[pdf]
[arXiv]
[bibtex]@InProceedings{Heitz_2021_CVPR, author = {Heitz, Eric and Vanhoey, Kenneth and Chambon, Thomas and Belcour, Laurent}, title = {A Sliced Wasserstein Loss for Neural Texture Synthesis}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, month = {June}, year = {2021}, pages = {9412-9420} }
A Sliced Wasserstein Loss for Neural Texture Synthesis
Abstract
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.
Related Material