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

2025-07-23

AI Summary

Text-to-image (T2I) models often suffer from detail misalignment and compositional distortion when generating images from prompts involving multiple subjects and complex attributes. To address this, we propose a **training-free progressive detail injection framework**: complex prompts are decomposed into sequential sub-prompts, and image generation proceeds in stages. Layout control is achieved via self-attention, while subject–attribute binding is enforced through cross-attention. Additionally, a test-time centroid alignment loss is introduced to refine the cross-attention mechanism, ensuring global compositional stability and precise fine-grained attribute binding. To our knowledge, this is the first method achieving multi-subject high-fidelity generation and consistent style control without any model training or fine-tuning. Extensive evaluation on T2I-CompBench and a newly constructed style-composition benchmark demonstrates significant improvements over state-of-the-art approaches—particularly in scenarios involving multi-object interactions and composite stylistic specifications.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Generation

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

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.

Problem

• Enhancing detail generation in text-to-image diffusion models
• Handling complex prompts with multiple distinct subjects
• Improving attribute binding and consistency in generated images

Innovation

• Progressive Detail Injection strategy
• Cross-attention mechanisms utilization
• Centroid Alignment Loss introduction