Perceptually Optimized Image Rendering

Valero Laparra, Alexander Berardino, Johannes Ballé, Eero Simoncelli

Abstract: We develop a framework for rendering photographic im- ages, taking into account display limitations, so as to op- timize perceptual similarity between the rendered image and the original scene. We formulate this as a constrained optimization problem, in which we minimize a measure of perceptual dissimilarity, the Normalized Laplacian Pyra- mid Distance (NLPD), which mimics the early stage trans- formations of the human visual system. When rendering images acquired with higher dynamic range than that of the display, we find that the optimized solution boosts the contrast of low-contrast features without introducing significant artifacts, yielding results of comparable visual quality to current state-of-the art methods with no man- ual intervention or parameter settings. We also examine a variety of other display constraints, including limitations on minimum luminance (black point), mean luminance (as a proxy for energy consumption), and quantized lu- minance levels (halftoning). Finally, we show that the method may be used to enhance details and contrast of images degraded by optical scattering (e.g., fog).