IEEE Project Abstract

High dynamic range (HDR) displays are capable of displaying a wider dynamic range of valuesthan conventional displays. As HDR content becomes more ubiquitous, the use of these displays is likely toaccelerate. As HDR displays can present a wider range of values, traditional strategies for mapping HDRcontent to low dynamic range (LDR) displays can be replaced with either directly displaying values, or usinga simple shift mapping (exposure adjustment). The latter approach is especially important when consideringambient lighting, as content viewed in a dark environment may appear substantially different to a bright one.This paper seeks to identify an exposure value which is suitable for displaying specific HDR content on anHDR display under a range of ambient lighting levels. Based on data captured with human participants, thispaper establishes user preferred exposure values for a variety of maximum display brightnesses, content andambient lighting levels. These are then used to develop two models to predict preferred exposure. The firstis based on linear regression using straightforward image statistics which require minimal computation andmemory to be computed, making this method suitable to be directly used in display hardware. The second isa model based on convolutional neural networks (CNN) to learn image features which best predict exposurevalues. The CNN model generates better results than the first model at the cost of memory and computationtime.