Detail++: Training-Free Detail Enhancer for Text-to-Image Diffusion Models

TL;DR

Detail++ is a training-free framework that improves text-to-image generation by decomposing complex prompts into stages, guiding global layout first and then refining details, especially for multiple subjects and styles.

Contribution

It introduces a novel Progressive Detail Injection strategy and a Centroid Alignment Loss, enabling staged, accurate, and style-consistent image synthesis without additional training.

Findings

Outperforms existing methods on T2I-CompBench

Enhances attribute binding accuracy

Improves handling of complex prompts with multiple objects

Abstract

Recent advances in text-to-image (T2I) generation have led to impressive visual results. However, these models still face significant challenges when handling complex prompt, particularly those involving multiple subjects with distinct attributes. Inspired by the human drawing process, which first outlines the composition and then incrementally adds details, we propose Detail++, a training-free framework that introduces a novel Progressive Detail Injection (PDI) strategy to address this limitation. Specifically, we decompose a complex prompt into a sequence of simplified sub-prompts, guiding the generation process in stages. This staged generation leverages the inherent layout-controlling capacity of self-attention to first ensure global composition, followed by precise refinement. To achieve accurate binding between attributes and corresponding subjects, we exploit cross-attention mechanisms and further introduce a Centroid Alignment Loss at test time to reduce binding noise and enhance attribute consistency. Extensive experiments on T2I-CompBench and a newly constructed style composition benchmark demonstrate that Detail++ significantly outperforms existing methods, particularly in scenarios involving multiple objects and complex stylistic conditions.