Mastering OpenCV With Practical Computer Vision Projects: Step-by-step Tutorials to Solve Common Real-world Computer Vision Problems for Desktop or Mobile, from Augmented Reality and Number Plate Recognition to Face Recognition and 3d Head Tracking

Of all the book reviews I’ve done this is the most genuine because I actually found this book useful for work I was doing. They found the right level of being technically interesting, robust and substantial all the while without being too daunting. The source code that accompanies this book is great and I still check back to it when starting new projects ([...]). Its well written and won’t take long to read through – I think it is a worthwhile read for anyone doing computer vision work. Program design isn’t something a lot of computer vision researchers / developers think or talk about a lot (at least in my experience) so to see how others lay out the problem can really help with your own work.

The book starts off by stating that it's intended for people with at least a bit of OpenCV knowledge and programming experience, and that rather than its aim is to show how to put together whole projects rather than explaining the theoretical background behind what you're doing. It's definitely correct in describing its target audience, and for the most part the focus is on implementation rather than theory. If you're looking for some examples of how to work with OpenCV as an all-around solution, this is a great collection for you.A big plus is that for the most part it uses the new OpenCV C++ interface. I say "mostly" because the authors also demonstrate projects for mobile devices - one for Android and one for iOS, which I really enjoyed reading.If you're looking for theory, you will find a bit in this book, but not much. I found the level that things were explained quite good, because you could understand what the code in the book does, but didn't necessarily have to be able to implement or understand the math behind them. The theoretical explanations have varying quality - sometimes they're right on the money, easy to understand and precisely relevant for what you're doing, and sometimes it's more than necessary. For the most part, I thought the authors did a good job of explaining what they were doing and why without forcing the reader to mire through lots of computer vision explanations that might not be interesting for somebody who's just wanting to implement a project that uses computer vision techniques. In every chapter, it was clear that the author really knew what he was talking about and many explanations you'll find in this book will be more concise and clear than what you'll find elsewhere.From a programming standpoint, the book is a bit of a potpourri. There are lots of really good tips for working with OpenCV in general and the authors always point you to further documetation. Also, besides the full example code used in the book, there are some nice utilities included that you might want to use for your own projects. A few chapters have very concise, clear code that's easy to follow and well-designed. In other places I felt that the code was much more quick and dirty. There's some using directives, which are a pet peeve for me, as well as really big try blocks, writing to globals, and unnecessary casting. In a lot of places, you'll notice the authors using the functions from the old C interface, even though they're writing in C++. In some places, you'll also find that the code is a bit redundant and difficult to extend, but that might actually be an advantage for a beginner who's more interested in seeing how to get something done than understanding complex but elegant design patterns. Taken as a whole, there is some nice code in the book, but there's also some very ugly code here too, depending on what part you're reading. It all implements the algorithms described in the chapters, however, and that is the focus of the book. It's definitely not a collection of best programming practices, but on the other hand, that's not what it's trying to be. You will find good recipes that contain lots of examples that you can apply to your own code, and they're explained well enough to make that easy for you too.I encourage you to read all the chapters of this book if you buy it. I read it with the intent of learning some techniques for my own projects, but although I could have done that by reading just a few chapters, I was glad to have read the whole thing. There are good tips and examples for working with OpenCV in general, and I found the sections on head pose estimation and face recognition very inspiring for my own work, even though I don't work with facial recognition at all. The book may not be a programming guide, but it definitely is a guide to computer vision and it shows you how to get very, very far, just by using OpenCV. It was well worth my time.

A person, totally unaware of CV or image processing have created an AR application for iOS for a week. IMHO, incredible.

I'm one of the authors of the book Practical Computer Vision with SimpleCV. The original reason we wrote the book was we felt OpenCV was lacking a lot of 'real world' type of examples that the average programmer could pick up without having to have a complex in-depth knowledge of Computer Vision. I feel this book does a very good job at that as well. Each chapter is basically it's own example with various computer vision techniques applied. I also appreciate the authors have posted all the code online for download and testing it the code compiled without any issues (Ubuntu 12.04). I definitely recommend this book if you are new to OpenCV or even interested in learning some of the basics of programming computer vision, although you should probably also have a bit of programming experience as well to actually understand what the code is doing.

The book contains a variety of different mini-projects, some simple, some extremely complicated. All are interesting and generally well presented. The only problem I have with this book is that the truly complex ones (such as chapter 6) could have used some deeper explanation of the math and algorithms used. Of course that would have meant the book had be either much bigger, or drop some of the chapters. In other words, it's a bit unclear if this is a book covering Computer Vision algorithms or OpenCV functionality - it tries to be both, and succeeds only partially.Bottom line - a good departure point for experienced OpenCV programmers who want to start their own projects.

This is the book I was looking for advanced projects with OpenCV 2. Really. It brings the best related to the most recent issues on face and character recognition.

Very Good Price

Really Good book.

One of the best book to learn opencv and computer vision. Highly recommended

Pretty good book - however, it does not come with "full example code" (quite a few bits - especially in chapters 6+7 - are left as "exercises for the reader") and the example code is not cross-platform compatible (allegedly, it works on Macs - but the "waitkey" calls need slight amendment for Linux) and quite a few people (including myself) have had trouble getting the example code working (it is all on Github - so it's worth checking the issues/push requests: the repo itself hasn't been updated since the book was published) - and quite a few links are outdated (I've informed Packt Publishing of these, but they've yet to publish the Errata despite acknowledging them). This book admits it isn't aimed at beginners (hence the "Mastering..." title - it expects you to have experience with OpenCV and C++ compiling), but some of the example are very "follow this simple code" and the guidance the text and code gives seems very "single usage" (i.e. not "You could then modify X here to do Y"). Slightly disappointing.

Wirklich gute Beispiele, hervorragendes Praxisbuch! Genau was man braucht um zu Arbeiten - keine überflüssigen Abschriften von Hilfedateien oder Funktionsbeschreibungen sondern praxisnahe Beispiele an denen man sich orientieren kann

First disapointment : the chapter 9 about using a kinect for natural body interface has just vanished. Though present in the cover, inside there is just no chapter 9. And there is no erratum about that for now.Second disapointment : the code in the book is often not up to date with the downloadable code though there was only one modification of the online code until now.In general, the scientific explanations in this book are really poor, there are rarely justifications about why using this method rather than this other method. Most of the time it is just "do like this", "do like that"; or even worse they tell you to do some method but then you realize that the given code goes another way.The content of this book may seem appealing but the realisation is really poor.Seems to me like a book made in a hurry for some reason, and not with time and the will to do good work.P.S. The first two comments are from an author and supposedly some of his friends, so don't take them into account.

Gives a good introduction to many parts of computer vision and how to implement them with opencv.Good source examples too.