On Theoretical Foundations of Human and Robot Vision

Abstract

Abstract A set of cognitive, neurological, and mathematical theories for human and robot vision has been recognized that encompasses David Hubel’s hypercolumn vision theory (The Nobel Prize in Physiology or Medicine 1981 [1]) and Dennis Gabor’s wavelet filter theory (The Nobel Prize in Physics 1971 [2]). This keynote lecture presents a theoretical framework of the Cognitive Vision Theory (CVT) [3-6] and its neurological and mathematical foundations. A set of Intelligent Mathematics (IM) [7-13] and formal vision theories developed in my laboratory is introduced encompassing Image Frame Algebra (IFA) [3], Visual Semantic Algebra (VSA) [4], and the Spike Frequency Modulation (SFM) theory [5]. IM is created for enabling cognitive robots to gain autonomous vision cognition capability supported by Visual Knowledge Bases (VKBs). Paradigms and case studies of robot vision powered by CVTs and IM will be demonstrated. The basic research on CVTs has led to new perspectives to human and robot vision for developing novel image processing applications in AI, neural networks, image recognitions, sequence learning, computational intelligence, self-driving vehicles, unmanned systems, and robot navigations.