DSI2I: Dense Style for Unpaired Image-to-Image Translation

TL;DR

DSI2I introduces a dense style representation for unpaired image-to-image translation, enabling finer style transfer without semantic labels, resulting in more diverse and accurate translations that better preserve content and match exemplars.

Contribution

The paper proposes a novel dense style representation for unpaired image translation, eliminating the need for semantic supervision and improving transfer quality.

Findings

More diverse translation outputs.

Better content preservation.

Closer style matching to exemplars.

Abstract

Unpaired exemplar-based image-to-image (UEI2I) translation aims to translate a source image to a target image domain with the style of a target image exemplar, without ground-truth input-translation pairs. Existing UEI2I methods represent style using one vector per image or rely on semantic supervision to define one style vector per object. Here, in contrast, we propose to represent style as a dense feature map, allowing for a finer-grained transfer to the source image without requiring any external semantic information. We then rely on perceptual and adversarial losses to disentangle our dense style and content representations. To stylize the source content with the exemplar style, we extract unsupervised cross-domain semantic correspondences and warp the exemplar style to the source content. We demonstrate the effectiveness of our method on four datasets using standard metrics together with a localized style metric we propose, which measures style similarity in a class-wise manner. Our results show that the translations produced by our approach are more diverse, preserve the source content better, and are closer to the exemplars when compared to the state-of-the-art methods. Project page: https://github.com/IVRL/dsi2i