FLOL: Fast Baselines for Real-World Low-Light Enhancement

  • 2025-01-16 18:06:09
  • Juan C. Benito, Daniel Feijoo, Alvaro Garcia, Marcos V. Conde
  • 0

Abstract

Low-Light Image Enhancement (LLIE) is a key task in computational photographyand imaging. The problem of enhancing images captured during night or in darkenvironments has been well-studied in the image signal processing literature.However, current deep learning-based solutions struggle with efficiency androbustness in real-world scenarios (e.g. scenes with noise, saturated pixels,bad illumination). We propose a lightweight neural network that combines imageprocessing in the frequency and spatial domains. Our method, FLOL+, is one ofthe fastest models for this task, achieving state-of-the-art results on popularreal scenes datasets such as LOL and LSRW. Moreover, we are able to process1080p images under 12ms. Code and models at https://github.com/cidautai/FLOL

 

Quick Read (beta)

loading the full paper ...