# Region-to-Region: Enhancing Generative Image Harmonization with Adaptive Regional Injection

arXiv: https://arxiv.org/abs/2508.09746 · HTML: https://mlsanity.com/paper/2508.09746/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 35 references that are in the map.
- First posted on arXiv 13 Aug 2025.

## Builds on (most cited first, 17 of 17)

- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [Teaching Large Language Models to Regress Accurate Image Quality Scores using Score Distribution](https://mlsanity.com/paper/2501.11561/) (2025)
- [DoveNet: Deep Image Harmonization via Domain Verification](https://mlsanity.com/paper/1911.13239/) (2019)
- [Deep Image Harmonization](https://mlsanity.com/paper/1703.00069/) (2017)
- [Improving the Harmony of the Composite Image by Spatial-Separated Attention Module](https://mlsanity.com/paper/1907.06406/) (2019)
- [Region-aware Adaptive Instance Normalization for Image Harmonization](https://mlsanity.com/paper/2106.02853/) (2021)
- [Making Images Real Again: A Comprehensive Survey on Deep Image Composition](https://mlsanity.com/paper/2106.14490/) (2021)
- [SSH: A Self-Supervised Framework for Image Harmonization](https://mlsanity.com/paper/2108.06805/) (2021)
- [Relightful Harmonization: Lighting-aware Portrait Background Replacement](https://mlsanity.com/paper/2312.06886/) (2023)
- [High-Resolution Image Harmonization via Collaborative Dual Transformations](https://mlsanity.com/paper/2109.06671/) (2021)
- [Foreground-aware Semantic Representations for Image Harmonization](https://mlsanity.com/paper/2006.00809/) (2020)
- [DCCF: Deep Comprehensible Color Filter Learning Framework for High-Resolution Image Harmonization](https://mlsanity.com/paper/2207.04788/) (2022)
- [Deep Image Harmonization by Bridging the Reality Gap](https://mlsanity.com/paper/2103.17104/) (2021)
- [Zero-Shot Image Harmonization with Generative Model Prior](https://mlsanity.com/paper/2307.08182/) (2023)
- [Learning Global-aware Kernel for Image Harmonization](https://mlsanity.com/paper/2305.11676/) (2023)
- [SIDNet: Learning Shading-aware Illumination Descriptor for Image Harmonization](https://mlsanity.com/paper/2112.01314/) (2021)

## Abstract (arXiv)

The goal of image harmonization is to adjust the foreground in a composite image to achieve visual consistency with the background. Recently, latent diffusion model (LDM) are applied for harmonization, achieving remarkable results. However, LDM-based harmonization faces challenges in detail preservation and limited harmonization ability. Additionally, current synthetic datasets rely on color transfer, which lacks local variations and fails to capture complex real-world lighting conditions. To enhance harmonization capabilities, we propose the Region-to-Region transformation. By injecting information from appropriate regions into the foreground, this approach preserves original details while achieving image harmonization or, conversely, generating new composite data. From this perspective, We propose a novel model R2R. Specifically, we design Clear-VAE to preserve high-frequency details in the foreground using Adaptive Filter while eliminating disharmonious elements. To further enhance harmonization, we introduce the Harmony Controller with Mask-aware Adaptive Channel Attention (MACA), which dynamically adjusts the foreground based on the channel importance of both foreground and background regions. To address the limitation of existing datasets, we propose Random Poisson Blending, which transfers color and lighting information from a suitable region to the foreground, thereby generating more diverse and challenging synthetic images. Using this method, we construct a new synthetic dataset, RPHarmony. Experiments demonstrate the superiority of our method over other methods in both quantitative metrics and visual harmony. Moreover, our dataset helps the model generate more realistic images in real examples. Our code, dataset, and model weights have all been released for open access.