Extreme Low-Light Image Denoising
Fig.1. Image capture setup and some sample images from our ELD datase
Extreme Low-light image denoising
Light is of paramount importance to photography. Night and low light place very demanding constraints on photography due to very limited photon count and inescapable noise. One natural reaction is to gather more light by, e.g., enlarging aperture setting, lengthening exposure time and opening flashlight. However, each method brings a trade-off–large aperture incurs small depth of field, and is usually unavailable in smartphone cameras; long exposure can induce blur due to scene variations or camera motions; flash can cause color aberrations and is useful only for nearby objects. Therefore, denoising in extremely low-light conditions has become an important research direction in the low-level image processing community, aiming to restore details in images captured in extremely low-light scenes and enhance visual quality.
We will use the Extreme Low-light image denoising dataset proposed by Prof. Fu’s team in [a]. We will host the competition using open source online platform, e.g. CodaLab. All submissions are evaluated by our script running on the server and we will double check the results of top-rank methods manually before releasing the final test-set rating.