BachGAN: High-Resolution Image Synthesis from Salient Object Layout

TL;DR

BachGAN introduces a novel approach for high-resolution image synthesis from simple object layouts, enabling realistic backgrounds and detailed textures without segmentation maps, advancing practical image generation applications.

Contribution

The paper presents BachGAN, a new model that generates high-quality images from object layouts by hallucinating backgrounds and seamlessly integrating foregrounds, addressing key challenges in layout-based image synthesis.

Findings

Outperforms existing methods in visual fidelity.

Produces images with better layout alignment.

Generates high-resolution, realistic images.

Abstract

We propose a new task towards more practical application for image generation - high-quality image synthesis from salient object layout. This new setting allows users to provide the layout of salient objects only (i.e., foreground bounding boxes and categories), and lets the model complete the drawing with an invented background and a matching foreground. Two main challenges spring from this new task: (i) how to generate fine-grained details and realistic textures without segmentation map input; and (ii) how to create a background and weave it seamlessly into standalone objects. To tackle this, we propose Background Hallucination Generative Adversarial Network (BachGAN), which first selects a set of segmentation maps from a large candidate pool via a background retrieval module, then encodes these candidate layouts via a background fusion module to hallucinate a suitable background for the given objects. By generating the hallucinated background representation dynamically, our model can synthesize high-resolution images with both photo-realistic foreground and integral background. Experiments on Cityscapes and ADE20K datasets demonstrate the advantage of BachGAN over existing methods, measured on both visual fidelity of generated images and visual alignment between output images and input layouts.