StarEnhancer: Learning Real-Time and Style-Aware Image Enhancement

TL;DR

StarEnhancer is a deep learning model capable of real-time, style-aware image enhancement across multiple styles, customizable by users, and effective on high-resolution images with superior quality metrics.

Contribution

The paper introduces a single, versatile model that handles multiple styles, supports user customization, and achieves high-speed processing for 4K images, surpassing existing methods.

Findings

Processes 4K images at over 200 FPS

Outperforms existing methods in PSNR, SSIM, LPIPS

Supports user-guided fine-tuning

Abstract

Image enhancement is a subjective process whose targets vary with user preferences. In this paper, we propose a deep learning-based image enhancement method covering multiple tonal styles using only a single model dubbed StarEnhancer. It can transform an image from one tonal style to another, even if that style is unseen. With a simple one-time setting, users can customize the model to make the enhanced images more in line with their aesthetics. To make the method more practical, we propose a well-designed enhancer that can process a 4K-resolution image over 200 FPS but surpasses the contemporaneous single style image enhancement methods in terms of PSNR, SSIM, and LPIPS. Finally, our proposed enhancement method has good interactability, which allows the user to fine-tune the enhanced image using intuitive options.