Adversarial Open Domain Adaptation for Sketch-to-Photo Synthesis

TL;DR

This paper introduces a novel open-domain sketch-to-photo translation framework that synthesizes realistic images from freehand sketches, even for classes absent in training data, by jointly learning sketch-photo mappings and employing a domain fooling strategy.

Contribution

It proposes a new open-domain adaptation method combining joint sketch-photo learning with a domain fooling strategy to handle missing class sketches.

Findings

Outperforms recent methods on Scribble and SketchyCOCO datasets.

Generates realistic color, texture, and preserves geometric structure.

Effective in synthesizing open-domain sketches into photos.

Abstract

In this paper, we explore open-domain sketch-to-photo translation, which aims to synthesize a realistic photo from a freehand sketch with its class label, even if the sketches of that class are missing in the training data. It is challenging due to the lack of training supervision and the large geometric distortion between the freehand sketch and photo domains. To synthesize the absent freehand sketches from photos, we propose a framework that jointly learns sketch-to-photo and photo-to-sketch generation. However, the generator trained from fake sketches might lead to unsatisfying results when dealing with sketches of missing classes, due to the domain gap between synthesized sketches and real ones. To alleviate this issue, we further propose a simple yet effective open-domain sampling and optimization strategy to "fool" the generator into treating fake sketches as real ones. Our method takes advantage of the learned sketch-to-photo and photo-to-sketch mapping of in-domain data and generalizes it to the open-domain classes. We validate our method on the Scribble and SketchyCOCO datasets. Compared with the recent competing methods, our approach shows impressive results in synthesizing realistic color, texture, and maintaining the geometric composition for various categories of open-domain sketches. Our code is available at https://github.com/Mukosame/AODA