Revisiting Sliced Wasserstein on Images: From Vectorization to ConvolutionDownload PDF

Published: 31 Oct 2022, 18:00, Last Modified: 23 Sept 2022, 14:33NeurIPS 2022 AcceptReaders: Everyone
Keywords: Sliced Wasserstein, Optimal Transport, Generative Models, Convolutional Operators
TL;DR: We propose convolution sliced Wasserstein between probability measures over images that are based on convolution operators.
Abstract: The conventional sliced Wasserstein is defined between two probability measures that have realizations as \textit{vectors}. When comparing two probability measures over images, practitioners first need to vectorize images and then project them to one-dimensional space by using matrix multiplication between the sample matrix and the projection matrix. After that, the sliced Wasserstein is evaluated by averaging the two corresponding one-dimensional projected probability measures. However, this approach has two limitations. The first limitation is that the spatial structure of images is not captured efficiently by the vectorization step; therefore, the later slicing process becomes harder to gather the discrepancy information. The second limitation is memory inefficiency since each slicing direction is a vector that has the same dimension as the images. To address these limitations, we propose novel slicing methods for sliced Wasserstein between probability measures over images that are based on the convolution operators. We derive \emph{convolution sliced Wasserstein} (CSW) and its variants via incorporating stride, dilation, and non-linear activation function into the convolution operators. We investigate the metricity of CSW as well as its sample complexity, its computational complexity, and its connection to conventional sliced Wasserstein distances. Finally, we demonstrate the favorable performance of CSW over the conventional sliced Wasserstein in comparing probability measures over images and in training deep generative modeling on images.
Supplementary Material: zip
40 Replies