InstaGAN: Instance-aware Image-to-Image Translation

TL;DR

InstaGAN is a novel image-to-image translation method that effectively handles multiple instances and significant shape changes by incorporating instance information and a context preserving loss.

Contribution

The paper introduces InstaGAN, which uses instance-aware features and a new loss to improve multi-instance translation in unsupervised GAN-based image translation.

Findings

Outperforms previous methods on challenging multi-instance translation tasks

Maintains permutation invariance of instances during translation

Generalizes well with limited GPU memory through sequential training

Abstract

Unsupervised image-to-image translation has gained considerable attention due to the recent impressive progress based on generative adversarial networks (GANs). However, previous methods often fail in challenging cases, in particular, when an image has multiple target instances and a translation task involves significant changes in shape, e.g., translating pants to skirts in fashion images. To tackle the issues, we propose a novel method, coined instance-aware GAN (InstaGAN), that incorporates the instance information (e.g., object segmentation masks) and improves multi-instance transfiguration. The proposed method translates both an image and the corresponding set of instance attributes while maintaining the permutation invariance property of the instances. To this end, we introduce a context preserving loss that encourages the network to learn the identity function outside of target instances. We also propose a sequential mini-batch inference/training technique that handles multiple instances with a limited GPU memory and enhances the network to generalize better for multiple instances. Our comparative evaluation demonstrates the effectiveness of the proposed method on different image datasets, in particular, in the aforementioned challenging cases. Code and results are available in https://github.com/sangwoomo/instagan