Supervisors: Dr. Radu Timofte
Although great success is achieved in recent years regarding the spectral reconstruction from RGB images, most of these methods assume known settings and/or a known mapping function to relate spectral data with the recorded RGB values. In contrast to the current literature, we address the problem of estimating the spectrum from a single common trichromatic RGB image obtained under unconstrained settings (e.g.unknown camera parameters, unknown scene radiance, unknown scene contents). For this we use a reference spectrum as provided by a hyperspectral image camera, and propose efficient deep learning solutions for sensitivity function estimation and spectral reconstruction from a single RGB image. We further expand the concept of spectral reconstruction such that to work for RGB images taken in the wild. We achieve state-of-the-art competitive results on the standard example-based spectral reconstruction benchmarks. Moreover, our experiments show that, for the first time, accurate spectral estimation from a single RGB image in the wild is within our reach.