FRIH: Fine-grained Region-aware Image Harmonization

TL;DR

FRIH introduces a two-stage, region-aware image harmonization framework that adaptively adjusts local regions for more realistic composite images, achieving state-of-the-art results with a lightweight model.

Contribution

The paper proposes a novel global-local two-stage framework for fine-grained region-aware image harmonization, improving detail preservation and regional adjustment over existing methods.

Findings

Achieves the best PSNR of 38.19 dB on iHarmony4 dataset.

Uses only 11.98 million parameters, significantly fewer than existing methods.

Demonstrates superior performance with a lightweight model.

Abstract

Image harmonization aims to generate a more realistic appearance of foreground and background for a composite image. Existing methods perform the same harmonization process for the whole foreground. However, the implanted foreground always contains different appearance patterns. All the existing solutions ignore the difference of each color block and losing some specific details. Therefore, we propose a novel global-local two stages framework for Fine-grained Region-aware Image Harmonization (FRIH), which is trained end-to-end. In the first stage, the whole input foreground mask is used to make a global coarse-grained harmonization. In the second stage, we adaptively cluster the input foreground mask into several submasks by the corresponding pixel RGB values in the composite image. Each submask and the coarsely adjusted image are concatenated respectively and fed into a lightweight cascaded module, adjusting the global harmonization performance according to the region-aware local feature. Moreover, we further designed a fusion prediction module by fusing features from all the cascaded decoder layers together to generate the final result, which could utilize the different degrees of harmonization results comprehensively. Without bells and whistles, our FRIH algorithm achieves the best performance on iHarmony4 dataset (PSNR is 38.19 dB) with a lightweight model. The parameters for our model are only 11.98 M, far below the existing methods.