Êtes-vous un étudiant de l'EPFL à la recherche d'un projet de semestre?
Travaillez avec nous sur des projets en science des données et en visualisation, et déployez votre projet sous forme d'application sur GraphSearch.
High bit depth data acquisition and manipulation have been largely studied at the academic level over the last 15 years and are rapidly attracting interest at the industrial level. An example of the increasing interest for high-dynamic range (HDR) imaging is the use of 32-bit floating point data for video and image acquisition and manipulation that allows a variety of visual effects that closely mimic the real-world visual experience of the end user [1] (see Figure 1). At the industrial level, we are witnessing increasing traction toward supporting HDR and wide color gamut (WCG). WCG leverages HDR for each color channel to display a wider range of colors. Consumer cameras are currently available with a 14- or 16-bit analog-to-digital converter. Rendering devices are also appearing with the capability to display HDR images and video with a peak brightness of up to 4,000 nits and to support WCG (ITU-R Rec. BT.2020 [2]) rather than the historical ITU-R Rec. BT.709 [3]. This trend calls for a widely accepted standard for higher bit depth support that can be seamlessly integrated into existing products and applications. While standard formats such as the Joint Photographic Experts Group (JPEG) 2000 [5] and JPEG XR [6] offer support for high bit depth image representations, their adoption requires a nonnegligible investment that may not always be affordable in existing imaging ecosystems, and induces a difficult transition, as they are not backward-compatible with the popular JPEG image format.