Artificial intelligence is part of our daily lives. How can we address its limitations and guide its use for the benefit of communities worldwide? Artificial intelligence (AI) has evolved from an experimental computer algorithm used by academic researchers to a commercially reliable method of sifting through large sets of data that detect patterns not readily apparent through more rudimentary search tools. As a result, AI-based programs are helping doctors make more informed decisions about patient care, city planners align roads and highways to reduce traffic congestion with better efficiency, and merchants scan financial transactions to quickly flag suspicious purchases. But as AI applications grow, concerns have increased, too, including worries about applications that amplify existing biases in business practices and about the safety of self-driving vehicles. In Can We Trust AI?, Dr. Rama Chellappa, a researcher and innovator with 40 years in the field, recounts the evolution of AI, its current uses, and how it will drive industries and shape lives in the future. Leading AI researchers, thought leaders, and entrepreneurs contribute their expertise as well on how AI works, what we can expect from it, and how it can be harnessed to make our lives not only safer and more convenient but also more equitable. Can We Trust AI? is essential reading for anyone who wants to understand the potential—and pitfalls—of artificial intelligence. The book features: • an exploration of AI's origins during the post–World War II era through the computer revolution of the 1960s and 1970s, and its explosion among technology firms since 2012; • highlights of innovative ways that AI can diagnose medical conditions more quickly and accurately; • explanations of how the combination of AI and robotics is changing how we drive; and • interviews with leading AI researchers who are pushing the boundaries of AI for the world's benefit and working to make its applications safer and more just. Johns Hopkins Wavelengths In classrooms, field stations, and laboratories in Baltimore and around the world, the Bloomberg Distinguished Professors of Johns Hopkins University are opening the boundaries of our understanding of many of the world's most complex challenges. The Johns Hopkins Wavelengths book series brings readers inside their stories, illustrating how their pioneering discoveries and innovations benefit people in their neighborhoods and across the globe in artificial intelligence, cancer research, food systems' environmental impacts, health equity, planetary science, science diplomacy, and other critical arenas of study. Through these compelling narratives, their insights will spark conversations from dorm rooms to dining rooms to boardrooms.
Face recognition has been actively studied over the past decade and continues to be a big research challenge. Just recently, researchers have begun to investigate face recognition under unconstrained conditions. Unconstrained Face Recognition provides a comprehensive review of this biometric, especially face recognition from video, assembling a collection of novel approaches that are able to recognize human faces under various unconstrained situations. The underlying basis of these approaches is that, unlike conventional face recognition algorithms, they exploit the inherent characteristics of the unconstrained situation and thus improve the recognition performance when compared with conventional algorithms. Unconstrained Face Recognition is structured to meet the needs of a professional audience of researchers and practitioners in industry. This volume is also suitable for advanced-level students in computer science.
Computer vision systems attempt to understand a scene and its components from mostly visual information. The geometry exhibited by the real world, the influence of material properties on scattering of incident light, and the process of imaging introduce constraints and properties that are key to solving some of these tasks. In the presence of noisy observations and other uncertainties, the algorithms make use of statistical methods for robust inference. In this paper, we highlight the role of geometric constraints in statistical estimation methods, and how the interplay of geometry and statistics leads to the choice and design of algorithms. In particular, we illustrate the role of imaging, illumination, and motion constraints in classical vision problems such as tracking, structure from motion, metrology, activity analysis and recognition, and appropriate statistical methods used in each of these problems.
Compressed sensing or compressive sensing is a new concept in signal processing where one measures a small number of non-adaptive linear combinations of the signal. These measurements are usually much smaller than the number of samples that define the signal. From these small numbers of measurements, the signal is then reconstructed by non-linear procedure. Compressed sensing has recently emerged as a powerful tool for efficiently processing data in non-traditional ways. In this book, we highlight some of the key mathematical insights underlying sparse representation and compressed sensing and illustrate the role of these theories in classical vision, imaging and biometrics problems.
This monograph is an outgrowth of the authors' recent research on the de velopment of algorithms for several low-level vision problems using artificial neural networks. Specific problems considered are static and motion stereo, computation of optical flow, and deblurring an image. From a mathematical point of view, these inverse problems are ill-posed according to Hadamard. Researchers in computer vision have taken the "regularization" approach to these problems, where one comes up with an appropriate energy or cost function and finds a minimum. Additional constraints such as smoothness, integrability of surfaces, and preservation of discontinuities are added to the cost function explicitly or implicitly. Depending on the nature of the inver sion to be performed and the constraints, the cost function could exhibit several minima. Optimization of such nonconvex functions can be quite involved. Although progress has been made in making techniques such as simulated annealing computationally more reasonable, it is our view that one can often find satisfactory solutions using deterministic optimization algorithms.
The recognition of humans and their activities from video sequences is currently a very active area of research because of its applications in video surveillance, design of realistic entertainment systems, multimedia communications, and medical diagnosis. In this lecture, we discuss the use of face and gait signatures for human identification and recognition of human activities from video sequences. We survey existing work and describe some of the more well-known methods in these areas. We also describe our own research and outline future possibilities. In the area of face recognition, we start with the traditional methods for image-based analysis and then describe some of the more recent developments related to the use of video sequences, 3D models, and techniques for representing variations of illumination. We note that the main challenge facing researchers in this area is the development of recognition strategies that are robust to changes due to pose, illumination, disguise, and aging. Gait recognition is a more recent area of research in video understanding, although it has been studied for a long time in psychophysics and kinesiology. The goal for video scientists working in this area is to automatically extract the parameters for representation of human gait. We describe some of the techniques that have been developed for this purpose, most of which are appearance based. We also highlight the challenges involved in dealing with changes in viewpoint and propose methods based on image synthesis, visual hull, and 3D models. In the domain of human activity recognition, we present an extensive survey of various methods that have been developed in different disciplines like artificial intelligence, image processing, pattern recognition, and computer vision. We then outline our method for modeling complex activities using 2D and 3D deformable shape theory. The wide application of automatic human identification and activity recognition methods will require the fusion of different modalities like face and gait, dealing with the problems of pose and illumination variations, and accurate computation of 3D models. The last chapter of this lecture deals with these areas of future research.
Thank you for visiting our website. Would you like to provide feedback on how we could improve your experience?
This site does not use any third party cookies with one exception — it uses cookies from Google to deliver its services and to analyze traffic.Learn More.