Haven’t read these 8 books, I am embarrassed to say that I am engaged in computer vision?
Programmer library (ID: OpenSourceTop) compile
Link: https://mc.ai/8-books-tartting-started-all-computer- vision/
Computer vision is a sub -field of artificial intelligence. It mainly focuses on digital images (such as photos and videos) and other aspects. Deep learning has made great progress in challenging computer visual tasks.
Before exploring the application of deep learning technology in computer vision, you need to have a computer visual basis, such as theoretical knowledge that master the foundation, commonly used technology, API standards and basic data processing
In this article, we will recommend a few popular computer vision -related top textbooks and computer visual books that are most suitable for programmers::
The best computer visual textbook
The textbook is written by experts (usually scholars), which aims to be a reference for students and practitioners. They mainly focus on general methods and theory (mathematics), not the application and method of actual problems (code).
According to their use in universities in top schools (such as MIT, etc.), the top five textbooks are listed:
1. “Computer Vision”
The PDF version of this book can be obtained for free on its official website
I like this book very much. It helps computer vision technology for beginners (undergraduates) to solve various basic computer vision problems. It is written based on the experience of teaching at the University of Washington for many years.
This book also reflects my 20 -year experience in computer vision research in the corporate research laboratory. Therefore, this book emphasizes the basic technologies in actual work, while paying less attention to deep mathematics with poor actual applicability.
2. “Computer Vision: Models, Learning, and Inference”
This book can still download the PDF version on its official website.
This is a good entry book (for students), covering a wide range of computer vision technology and problems. This book takes more lengths to introduce computer vision, and spend a lot of space on the basic theme of probability modeling
This book pays more attention to the probability model of learning and reasoning. It shows how to use training data to learn the relationship between image data observed and our real world, such as 3D structures or object categories. The foundation starts until the reader can realize and modify the real example to build a useful visual system. Mainly for senior undergraduate and graduate students,
3. “Computer Vision: A Modern Method”
This is an entry textbook about computer vision. Its theme may be wider than many other textbooks. Despite the extensive content, it may not be as popular as other introductory texts because it may be more challenging to the narrative method directly into the theme.
The content of this book involves geometric camera models, light and coloring, color, linear filtering, local image characteristics, texture, three -dimensional relative, motion structure, cluster division, combination and model fitting, tracking, standard, smooth surface and skeleton, distance, distance Data, image classification, object detection and identification, image -based modeling and rendering, humanoid research, image search and retrieval, optimization technology and other content.
4. “Introiductory Techniques for 3-D Computer Vision”
This is an older book. It mainly focuses on computer vision. There are also some technologies that are concerned about 3D vision. It is a practical introduction to modern computer vision problems and solutions. This is a good starting point, suitable for readers with undergraduate rather than graduate level.
5. “Multiple View Geometry in Computer Vision”
Some chapters of this book can download the PDF version from the official website
This book is quite advanced in computer vision, and it is suitable for people who read the level of graduate students, especially the problems and methods of inferring geometry from multiple images.
This book is divided into six parts and seven short appendixs. Some people say that if you have n’t read computer vision, you are embarrassed to say that you are engaged in computer vision
Computer visual books suitable for programmers
Programmers’ books are written by experts (usually developers and engineers) (such as O’Reilly books). They mainly focus on practical issues of technology and solving problems, and focus on sample code and standard library. Technology can be briefly described in the relevant theory (mathematics), but it is not used for the main reference.
6. “Learning OPENCV (Chinese Version)”
The focus of this book is to teach you how to use the OpenCV library, which may be the first open source computer vision library. All code examples are written in C ++. More importantly, the author is a member and founder of the board of directors of OpenCV.
It is a technical book, but it is like a carefully written API document. This book provides a guide to use the C ++ open source computer vision library OpenCV 3. It also gives the basic background knowledge in the field of computer vision to help readers use OpenCV more effectively.
7. “Python Computer Visual Programming”
This is a practical book, which mainly teaches you how to use Python to perform basic computer visual tasks, mainly using Pil, but also about OpenCV’s basic introduction.
There are a lot of examples in the book carefully analyzed object recognition, content -based image search, optical character recognition, light flow method, tracking, three -dimensional reconstruction, three -dimensional imaging, enhanced reality, gesture estimation, panoramic creation, image segmentation, noise reduction, image grouping Waiting for technology.
8. “Practical Computer Vision with Simplecv”
This book teaches you how to use the SimpleCV Library in Python to execute the basic computer visual operation. SimpleCV is an open source framework written in Python, which is a good alternative to Pil (Pillow)/OpenCV.
Through practical examples, this book teaches you how to use SimpleCV to quickly and easily build your own computer vision (CV) application, and at the same time introduce you to the basic CV technology of collecting, processing and analyzing streaming digital images.