Learning to Enhance Visual Quality via Hyperspectral Domain Mapping
This work addresses low-light image enhancement for computer vision applications, but it is incremental as it builds on existing unpaired cycle-consistent frameworks.
The paper tackles low-light image enhancement by generating hyperspectral images from RGB inputs to estimate pixel-wise dynamic range adjustments, achieving improved visual quality on the LOL Dataset.
Deep learning based methods have achieved remarkable success in image restoration and enhancement, but most such methods rely on RGB input images. These methods fail to take into account the rich spectral distribution of natural images. We propose a deep architecture, SpecNet, which computes spectral profile to estimate pixel-wise dynamic range adjustment of a given image. First, we employ an unpaired cycle-consistent framework to generate hyperspectral images (HSI) from low-light input images. HSI is further used to generate a normal light image of the same scene. We incorporate a self-supervision and a spectral profile regularization network to infer a plausible HSI from an RGB image. We evaluate the benefits of optimizing the spectral profile for real and fake images in low-light conditions on the LOL Dataset.