Color is an essential feature of objects for a wide range of applications, such as digital photography,content-based image retrieval, object tracking/recognition, image forensics, human-computer interactionand augmented reality. Human Vision System has the remarkable capability to perceive the color of theobjects in the scene mostly invariable in spite of changes in the illumination/imaging conditions. However,it is not equally simple for a machine vision algorithm to alleviate the influence of these conditions makingthe computer able to see like we do. In fact, when we acquire and image, the recorded color of an object isnot an intrinsic property of the object itself, but it depends also on the (often unknown) illuminationconditions under which the object has been acquired. Computational Color Constancy algorithms aim toestimate the actual color in an acquired scene disregarding its illuminant, rendering the color of the objectsin the scene invariable in spite of changes in the illumination conditions. It can be shown that colorconstancy is an ill-posed problem, its solution lacks therefore of uniqueness and stability. To cope with thisproblem, common solutions usually exploit some heuristic assumptions about the statistical properties ofthe expected illuminants and/or of the reflectance of the objects in the scene. In this seminar I brieflyreview state-of- the-art methods and illustrate promising researches aimed to improve illuminantestimation by using visual and/or semantic information automatically extracted from the images.
Raimondo Schettini is a full professor at the University of Milano-Bicocca (Italy). He is Vice-Director of the Department of Informatics, Systems and Communication, and head of Imaging and Vision Lab (www.ivl.disco.unimib.it). He has been associated with Italian National Research Council (CNR) since 1987 where he has leaded the Color Imaging lab from 1990 to 2002. He has been team leader in several research projects and published more than 300 papers and several patents about color imaging, image processing, analysis and classification, image and video indexing and retrieval. Raimondo is a fellow of the International Association of Pattern Recognition (IAPR) for his contributions to pattern recognition research and color image analysis.