Michael S. Landy – författare

This book provides an introduction into both computational models and experimental paradigms that are concerned with sensory cue integration both within and between sensory modalities. Importantly, across behavioral, electrophysiological and theoretical approaches, Bayesian statistics is emerging as a common language in which cue-combination problems can be expressed. This book focuses on the emerging probabilistic way of thinking about these problems. These approaches derive from the realization that all our sensors are noisy and moreover are often affected by ambiguity. For example, mechanoreceptor outputs are variable and they cannot distinguish if a perceived force is caused by the weight of an object or by force we are producing ourselves. The computational approaches described in this book aim at formalizing the uncertainty of cues. They describe cue combination as the nervous system's attempt to minimize uncertainty in its estimates and to choose successful actions. Some computational approaches described in the chapters of this book are concerned with the application of such statistical ideas to real-world cue-combination problems, such as shape and depth perception. Other parts of the book ask how uncertainty may be represented in the nervous system and used for cue combination. The broadening scope of probabilistic approaches to cue combination is highlighted in the breadth of topics covered in this book: the chapters summarize and discuss computational approaches and behavioral evidence aimed at understanding the combination of visual, auditory, proprioceptive, and haptic cues. Some chapters address the combination of cues within a single sensory modality while others address the combination across sensory modalities. Neural implementation, behavior, and theory are considered. The unifying aspect of this book is the focus on the uncertainty intrinsic to sensory cues and the underlying question of how the nervous system deals with this uncertainty.The book is intended as a reference text for graduate students and professionals in perceptual psychology, computational neuroscience, cognitive neuroscience and sensory neurophysiology.

This book is concerned with sensory cue integration both within and between sensory modalities, and focuses on the emerging way of thinking about cue combination in terms of uncertainty. These probabilistic approaches derive from the realization that our sensors are noisy and moreover are often affected by ambiguity. For example, mechanoreceptor outputs are variable and they cannot distinguish if a perceived force is caused by the weight of an object or by force we are producing ourselves. The probabilistic approaches elaborated in this book aim at formalizing the uncertainty of cues. They describe cue combination as the nervous system''s attempt to minimize uncertainty in its estimates and to choose successful actions. Some computational approaches described in the chapters of this book are concerned with the application of such statistical ideas to real-world cue-combination problems. Others ask how uncertainty may be represented in the nervous system and used for cue combination. Importantly, across behavioral, electrophysiological and theoretical approaches, Bayesian statistics is emerging as a common language in which cue-combination problems can be expressed.

This book is a dialogue between researchers who study biological visual and computer scientists and engineers who seek to build computer vision systems that actively explore the environment. By describing new and important ways to design robots analogous to biological visual systems, it provides deep insights into the problems and solutions of computer vision. The book is divided into four parts, each addressing a different aspect of exploratory or active vision in biological and machine vision systems. The chapters are written by a cross-disciplinary selection of leading researchers who study computer and biological vision. As a result, many researchers and students concerned with vision will find this an invaluable survey to this fast-moving field.

Advances in sensing, signal processing, and computer technology during the past half century have stimulated numerous attempts to design general-purpose ma­ chines that see. These attempts have met with at best modest success and more typically outright failure. The difficulties encountered in building working com­ puter vision systems based on state-of-the-art techniques came as a surprise. Perhaps the most frustrating aspect of the problem is that machine vision sys­ tems cannot deal with numerous visual tasks that humans perform rapidly and effortlessly. In reaction to this perceived discrepancy in performance, various researchers (notably Marr, 1982) suggested that the design of machine-vision systems should be based on principles drawn from the study of biological systems. This "neuro­ morphic" or "anthropomorphic" approach has proven fruitful: the use of pyramid (multiresolution) image representation methods in image compression is one ex­ ample of a successful application based on principles primarily derived from the study of biological vision systems. It is still the case, however, that the perfor­ of computer vision systems falls far short of that of the natural systems mance they are intended to mimic, suggesting that it is time to look even more closely at the remaining differences between artificial and biological vision systems.