Evaluating Low-Light Image Enhancement Across Multiple Intensity Levels
This work addresses the lack of radiance diversity in low-light enhancement evaluation, providing a benchmark for researchers and practitioners in computer vision and imaging.
The paper tackled the problem of evaluating low-light image enhancement methods across varying illumination intensities by introducing the Multi-Illumination Low-Light (MILL) dataset, which revealed significant performance variations and led to improvements achieving up to 10 dB PSNR gain for DSLR and 2 dB for smartphone images.
Imaging in low-light environments is challenging due to reduced scene radiance, which leads to elevated sensor noise and reduced color saturation. Most learning-based low-light enhancement methods rely on paired training data captured under a single low-light condition and a well-lit reference. The lack of radiance diversity limits our understanding of how enhancement techniques perform across varying illumination intensities. We introduce the Multi-Illumination Low-Light (MILL) dataset, containing images captured at diverse light intensities under controlled conditions with fixed camera settings and precise illuminance measurements. MILL enables comprehensive evaluation of enhancement algorithms across variable lighting conditions. We benchmark several state-of-the-art methods and reveal significant performance variations across intensity levels. Leveraging the unique multi-illumination structure of our dataset, we propose improvements that enhance robustness across diverse illumination scenarios. Our modifications achieve up to 10 dB PSNR improvement for DSLR and 2 dB for the smartphone on Full HD images.