Multi-Channel Attention Selection GANs for Guided Image-to-Image Translation

TL;DR

SelectionGAN is a novel two-stage model for guided image-to-image translation that leverages multi-channel attention and uncertainty-guided pixel loss to produce superior results across various challenging tasks.

Contribution

It introduces a multi-stage framework with attention modules and uncertainty-guided loss, advancing the state-of-the-art in guided image translation.

Findings

Outperforms existing methods on face, hand, body, and street view translation tasks.

Utilizes attention maps to learn uncertainty for improved pixel loss.

Framework adaptable to other image generation tasks.

Abstract

We propose a novel model named Multi-Channel Attention Selection Generative Adversarial Network (SelectionGAN) for guided image-to-image translation, where we translate an input image into another while respecting an external semantic guidance. The proposed SelectionGAN explicitly utilizes the semantic guidance information and consists of two stages. In the first stage, the input image and the conditional semantic guidance are fed into a cycled semantic-guided generation network to produce initial coarse results. In the second stage, we refine the initial results by using the proposed multi-scale spatial pooling & channel selection module and the multi-channel attention selection module. Moreover, uncertainty maps automatically learned from attention maps are used to guide the pixel loss for better network optimization. Exhaustive experiments on four challenging guided image-to-image translation tasks (face, hand, body, and street view) demonstrate that our SelectionGAN is able to generate significantly better results than the state-of-the-art methods. Meanwhile, the proposed framework and modules are unified solutions and can be applied to solve other generation tasks such as semantic image synthesis. The code is available at https://github.com/Ha0Tang/SelectionGAN.