Computer Vision Metrics: Survey, Taxonomy, and Analysis of Computer Vision, Visual Neuroscience, and Visual AI

ISBN-13: 9789819933921 / Angielski

Scott Krig

Spis treści

Chapter 1. 2D/3D Image Capture and Representation.- Chapter 2. Image Pre-Processing Taxonomy, Colorimetry.- Chapter 3. Global and Regional Feature Descriptors.- Chapter 4. Local Feature Descriptors.- Chapter 5. Feature Descriptor Attribute Taxonomy.- Chapter 6. Feature Detector and Descriptor Survey.- Chapter 7. Ground Truth Data Topics, Benchmarks, Analysis.- Chapter 8. Vision Pipelines and HW/SW Optimizations.- Chapter 9. Feature Learning Taxonomy and Neuroscience Background.-Chapter 10. Feature Learning and Deep Learning Survey.- Chapter 11. Attention, Transformers, Hybrids, DDN’s.- Chapter 12. Applied And Future Visual Computing Topics.

O autorze

Scott Krig is a pioneer in computer imaging, computer vision, and graphics visualization. He founded Krig Research in 1988, providing the world’s first image and vision systems based on high-performance engineering workstations, super-computers, and dedicated hardware, with optimized computer vision and imaging software libraries for a wide range of applications, serving customers in 25 countries around the globe. Scott is also the author of Synthetic Vision Using Volume Learning and Visual DNA, which presents a multi-dimensional and multivariate feature learning approach to computer vision, intended as the basis for a public Visual Genome Project to catalog all (or nearly all) visual features composing visual objects. Scott studied at Stanford and is the author of patent applications worldwide in various fields, including imaging, computer vision, embedded systems, DRM and computer security. 

Blurb

This 2nd Edition, based on the successful 2016 textbook, has been updated and expanded to cover 3rd generation Computer Vision and AI as it supersedes historical visual computing methods, providing a comprehensive survey of essential topics and methods in Computer Vision. With over 1,200 essential references, as well as chapter-by-chapter learning assignments, the book offers a valuable resource for students, researchers, scientists and engineers, helping them dig deeper into core computer vision and foundational visual computing and neuroscience topics. 

As before, a historical survey of advances in Computer Vision is provided, updated to reflect the latest methods such as Vision Transformers, attention models, alternative features such as Fourier neurons and Binary neurons, hybrid DNN architectures, self-supervised and enhanced learning models, Associative Multimodal Learning, Continuous Learning, View Synthesis, intelligent Scientific Imaging, and advances in training protocols. Updates have also been added for 2d/3d cameras, software libraries and open source resources, computer vision cloud services, and vision/AI hardware accelerators. Discussion and analysis are provided to uncover intuition and delve into the essence of key advancements, applied and forward-looking topics.