Artificial intelligence (AI) has continued to be a buzzword in recent years mainly due to the advances of deep learning techniques and the call from big data. The basic definition of AI requires the computer to see, to understand, and to make decisions about the environment. That is why computer vision is by nature part of AI. However, we should not forget that the purpose of AI is to help humans accomplish tasks. If you ever wonder with what areas AI can help, take a look at the index of this book.
A few very different applications are discussed in the book’s 11 chapters, namely ocean observing, fault diagnosis/classification, visual cryptography, weld defect extraction/evaluation, image retrieval, public opinion detection, image compression, human action recognition, and light field vision. Some of them are relatively old topics, such as image compression, but a few are quite interesting in today’s hot market. Read chapter 11 if you want to catch up with the latest developments in virtual/augmented reality, which are backed up by light field vision. Chapter 9 is very practical if you are interested in building a video-based action detection system, which can be used to collect valuable user behavior data.
In fact, most of the chapters (except for chapter 1) contain lots of implementation details and experimental data. Thus, this volume could be used as a supplemental textbook. The text is easy to read and understand; however, some figures could be printed in better quality, especially in chapter 3. I wouldn’t recommend the book to readers who are looking for the latest techniques--for example, deep neural networks are not included. The bottom line: the book will provide you with ideas about different AI-driven projects.