RegionCache: Semantic-Aware Region Reuse for Efficient Multi-Turn Image Generation

2026-08-30

AI Summary

针对多轮图像编辑中的冗余计算问题，提出RegionCache框架，通过语义感知选择性重用未改变区域的扩散状态，提高效率并保持图像质量。

Abstract

Real-world image generation often involves multi-turn editing, where users iteratively modify small regions while most image content remains unchanged. However, existing diffusion transformer (DiT)-based editing pipelines recompute the entire image at every turn, causing substantial redundant computation. Existing DiT acceleration methods further ignore semantic correspondence across prompts, leading to unnecessary recomputation or unsafe reuse that harms editing quality. To address this, we propose RegionCache, a semantic-aware reuse framework for multi-turn image editing that selectively reuses diffusion states from unchanged regions. RegionCache detects reusable regions through semantic overlap between consecutive prompts and cross-attention localization, and adopts an adaptive reuse schedule based on prompt similarity and contextual consistency. Experiments on PixArt-alpha demonstrate that RegionCache achieves 1.43x--2.55x end-to-end speedup while maintaining comparable image quality. Code is available at https://github.com/hebutBryant/RegionCache.

Problem

• multi-turn image generation
• diffusion transformer
• redundant computation
• semantic correspondence
• editing quality

Innovation

• Semantic-Aware Reuse
• Multi-Turn Image Editing
• Diffusion States
• Cross-Attention Localization
• Adaptive Reuse Schedule