Infrared Colorization.zip

Dataset used in Browne, Deyneka, Ceccarelli, Chen, To, Tang, Vu, Baldi 'Deep Learning to Enable Color Vision in the Dark' <br>Description of the dataset:Labeled Faces in the Wild: A Database for Studying Face Recognition in Unconstrained Environments, illuminated with diffuse uniform illumination by 6 different LED wavelengths spanning the visible and near-infrared spectrum (center wavelengths: 447, 529, 604, 718, 777, 807nm)Number of samples: 200Illuminations: 447, 529, 604, 718, 777, 807 nm + merged-colour RGB imageImage size: (2056, 2464)<br><br>Photographs of each printed image under multispectral illumination were acquired using a monochromatic camera (Zeiss Axiocam 305, CarlZeiss) mounted on a dissection microscope focused on the image. To account for diminishing camera quantum efficiency in the red and infrared spectrum, each image channel was normalized to the white background of the paper upon which the images were printed. <br>The dataset was used for the development of machine learning models to predict RGB color images from individual or combinations of wavelength illuminations.<br>