Attention-Guided Generative Adversarial Networks for Unsupervised Image-to-Image Translation

Abstract

The state-of-the-art approaches in Generative Adversarial Networks (GANs) are able to learn a mapping function from one image domain to another with unpaired image data. However, these methods often produce artifacts and can only be able to convert low-level information, but fail to transfer high-level semantic part of images. The reason is mainly that generators do not have the ability to detect the most discriminative semantic part of images, which thus makes the generated images with low-quality. To handle the limitation, in this paper we propose a novel Attention-Guided Generative Adversarial Network (AGGAN), which can detect the most discriminative semantic object and minimize changes of unwanted part for semantic manipulation problems without using extra data and models. The attention-guided generators in AGGAN are able to produce attention masks via a built-in attention mechanism, and then fuse the input image with the attention mask to obtain a target image with high-quality. Moreover, we propose a novel attention-guided discriminator which only considers attended regions. The proposed AGGAN is trained by an end-to-end fashion with an adversarial loss, cycle-consistency loss, pixel loss and attention loss. Both qualitative and quantitative results demonstrate that our approach is effective to generate sharper and more accurate images than existing models. The code is available at https://github.com/Ha0Tang/AttentionGAN.

8 pages, 7 figures, Accepted to IJCNN 2019

References in corpus (12)

• Conditional Generative Adversarial Nets
• Large Scale GAN Training for High Fidelity Natural Image Synthesis
• Self-Attention Generative Adversarial Networks
• Toward Multimodal Image-to-Image Translation
• Learning to Discover Cross-Domain Relations with Generative Adversarial Networks
• Invertible Conditional GANs for image editing
• DualGAN: Unsupervised Dual Learning for Image-to-Image Translation
• Unsupervised Attention-guided Image to Image Translation
• Learning What and Where to Draw
• Mask-aware Photorealistic Face Attribute Manipulation
• Generative Semantic Manipulation with Contrasting GAN
• Attention-GAN for Object Transfiguration in Wild Images

Cited by in corpus (1)

• Total Generate: Cycle in Cycle Generative Adversarial Networks for Generating Human Faces, Hands, Bodies, and Natural Scenes