ST-GAN: Spatial Transformer Generative Adversarial Networks for Image Compositing
Abstract
We address the problem of finding realistic geometric corrections to a foreground object such that it appears natural when composited into a background image. To achieve this, we propose a novel Generative Adversarial Network (GAN) architecture that utilizes Spatial Transformer Networks (STNs) as the generator, which we call Spatial Transformer GANs (ST-GANs). ST-GANs seek image realism by operating in the geometric warp parameter space. In particular, we exploit an iterative STN warping scheme and propose a sequential training strategy that achieves better results compared to naive training of a single generator. One of the key advantages of ST-GAN is its applicability to high-resolution images indirectly since the predicted warp parameters are transferable between reference frames. We demonstrate our approach in two applications: (1) visualizing how indoor furniture (e.g. from product images) might be perceived in a room, (2) hallucinating how accessories like glasses would look when matched with real portraits.
Cite
@article{arxiv.1803.01837,
title = {ST-GAN: Spatial Transformer Generative Adversarial Networks for Image Compositing},
author = {Chen-Hsuan Lin and Ersin Yumer and Oliver Wang and Eli Shechtman and Simon Lucey},
journal= {arXiv preprint arXiv:1803.01837},
year = {2018}
}
Comments
Accepted to CVPR 2018 (website & code: https://chenhsuanlin.bitbucket.io/spatial-transformer-GAN/)