A Unified Approach to Data Analysis and Modeling of the Appearance of Materials for Computer Graphics and Multidimensional Reflectometry

摘要: Characterizing the appearance of real-world surfaces is a fundamental problem in multidimensional , computer vision and graphics. In this paper, we outline unified perception-based approach to modeling materials for graphics reflectometry. We discuss differences common points data analysis BRDFs both physical virtual application domains. mathematical framework that captures important problems types domains, allows performance comparisons statistical machine learning methods. For between methods, use criteria are relevant statistics learning, as well Additionally, propose class multiple testing procedures test hypothesis material has diffuse reflection generalized sense. treat general case where number hypotheses can potentially grow with measurements. Our leads tests more powerful than generic procedures.