Few-Shot Unsupervised Image-to-Image Translation on complex scenes

TL;DR

This paper evaluates and extends the FUNIT unsupervised image-to-image translation method for complex, content-rich scenes by incorporating object detection and diverse datasets to improve realism and applicability.

Contribution

It adapts the FUNIT framework to handle complex scenes using object detection and diverse datasets, expanding its applicability beyond single object translation.

Findings

FUNIT performs reasonably on complex scenes with modifications.

Object detection integration improves translation quality.

Diverse datasets help in understanding method behavior on complex images.

Abstract

Unsupervised image-to-image translation methods have received a lot of attention in the last few years. Multiple techniques emerged tackling the initial challenge from different perspectives. Some focus on learning as much as possible from several target style images for translations while other make use of object detection in order to produce more realistic results on content-rich scenes. In this work, we assess how a method that has initially been developed for single object translation performs on more diverse and content-rich images. Our work is based on the FUNIT[1] framework and we train it with a more diverse dataset. This helps understanding how such method behaves beyond their initial frame of application. We present a way to extend a dataset based on object detection. Moreover, we propose a way to adapt the FUNIT framework in order to leverage the power of object detection that one can see in other methods.