Spatial-Separated Curve Rendering Network for Efficient and High-Resolution Image Harmonization

TL;DR

The paper introduces S$^2$CRNet, a lightweight and efficient neural network for high-resolution image harmonization that significantly reduces parameters and computational cost while maintaining state-of-the-art performance.

Contribution

Proposes a novel spatial-separated curve rendering network (S$^2$CRNet) for efficient high-resolution image harmonization, with extensions for cascaded refinement and semantic guidance.

Findings

Reduces over 90% of parameters compared to previous methods.

Achieves state-of-the-art performance on iHarmony4 and DIH datasets.

Operates on 2048x2048 images in 0.1 seconds with low GPU resources.

Abstract

Image harmonization aims to modify the color of the composited region with respect to the specific background. Previous works model this task as a pixel-wise image-to-image translation using UNet family structures. However, the model size and computational cost limit the ability of their models on edge devices and higher-resolution images. To this end, we propose a novel spatial-separated curve rendering network(S$^2$CRNet) for efficient and high-resolution image harmonization for the first time. In S$^2$CRNet, we firstly extract the spatial-separated embeddings from the thumbnails of the masked foreground and background individually. Then, we design a curve rendering module(CRM), which learns and combines the spatial-specific knowledge using linear layers to generate the parameters of the piece-wise curve mapping in the foreground region. Finally, we directly render the original high-resolution images using the learned color curve. Besides, we also make two extensions of the proposed framework via the Cascaded-CRM and Semantic-CRM for cascaded refinement and semantic guidance, respectively. Experiments show that the proposed method reduces more than 90% parameters compared with previous methods but still achieves the state-of-the-art performance on both synthesized iHarmony4 and real-world DIH test sets. Moreover, our method can work smoothly on higher resolution images(eg., $2048\times2048$) in 0.1 seconds with much lower GPU computational resources than all existing methods. The code will be made available at \url{http://github.com/stefanLeong/S2CRNet}.