LSReGen: Large-Scale Regional Generator via Backward Guidance Framework

TL;DR

This paper introduces LSReGen, a novel large-scale layout-to-image generation framework that improves controllability and quality by generalizing backward guidance, outperforming existing methods in large-scale tasks.

Contribution

The paper presents a new backward guidance framework and LSReGen, enabling high-quality, layout-controlled image generation at large scale with better performance than prior approaches.

Findings

LSReGen achieves superior quality in large-scale layout-to-image tasks.

The framework generalizes backward guidance without model-specific assumptions.

Experimental results outperform existing methods.

Abstract

In recent years, advancements in AIGC (Artificial Intelligence Generated Content) technology have significantly enhanced the capabilities of large text-to-image models. Despite these improvements, controllable image generation remains a challenge. Current methods, such as training, forward guidance, and backward guidance, have notable limitations. The first two approaches either demand substantial computational resources or produce subpar results. The third approach depends on phenomena specific to certain model architectures, complicating its application to large-scale image generation.To address these issues, we propose a novel controllable generation framework that offers a generalized interpretation of backward guidance without relying on specific assumptions. Leveraging this framework, we introduce LSReGen, a large-scale layout-to-image method designed to generate high-quality, layout-compliant images. Experimental results show that LSReGen outperforms existing methods in the large-scale layout-to-image task, underscoring the effectiveness of our proposed framework. Our code and models will be open-sourced.