Computer vision applications are changing the world. Allowing computers to understand and label images has brought us (closer) to driverless driving, face-based access control, social robots, automated medical diagnostics… and countless simple, everyday apps such as Snapchat.
Computer vision has become a truly exciting field to work in. If you’d like to explore it, reading is the best way to build a strong foundation for any practical projects you may take on in the future. This article presents our top picks among computer vision books – hopefully, it will help you choose your next great read.
Author: Richard Szeliski
Richard Szeliski is a well-known name in computer vision circles and so is his book. Although it was primarily designed as a students’ textbook, Computer Vision: Algorithms and Applications has become a great reference point for all computer vision enthusiasts – students, researchers and professionals alike.
Szeliski’s book explores the variety of techniques for analyzing and interpreting images illustrated by hundreds of color photos. It not only covers all the fundamentals of computer vision but also recommends the latest research literature. Besides teaching the theory, Szeliski does his best to encourage the readers to try out their new knowledge on real-world projects by emphasizing the importance of testing algorithms and presenting useful techniques that work under real-world conditions.
If you want a comprehensive and clear introduction to computer vision, this book is a great start. However, since it covers so much information, you should consider it a high-level introduction and a starting point for learning more.
Authors: Richard Hartley and Andrew Zisserman
Multiple View Geometry in Computer Vision is one of the fundamental computer vision books – it’s even considered the bible of computer vision by many researchers. The book explains the geometry behind 3D reconstructions, which is essential in fusing information gained from camera movement.
The book is extremely well written and informative. The authors break down complex algorithms into simple English, explaining how to successfully implement them. It’s a must have for anyone working on 3D computer vision. The book may be challenging for beginners, who might be better off starting with Trucco’s Introductory Techniques for 3-D Computer Vision.
Author: Simon J. D. Prince
Computer Vision – Models, Learning and Inference treats computer vision from a machine learning point of view. It explains the basics of probability and model fitting and builds up to cutting-edge techniques and valuable examples the readers can then implement themselves.
The book describes more than 70 algorithms intuitively, so the reader can easily implement them on their own. The material is well-organized and comes with hundreds of excellent, full-color illustrations. Overall, it’s a great introductory book that will lay a solid foundation for future learning.
Authors: Ian Goodfellow, Yoshua Bengio, Aaron Courville
Deep Learning is THE book on deep learning that discusses a broad range of topics in the field. It covers the machine learning basics, modern deep learning algorithms and techniques and, finally, the latest research trends and perspectives.
Since it was primarily written for an academic audience, Deep Learning is entirely theoretical, so you won’t find any code covered in the book. Instead, you can expect to dive into practically applicable yet high level theory. However, it provides extremely valuable knowledge on deep learning that will come handy in your future work, be it industry or research.
… said Epictetus hundreds of years ago and it couldn’t be more true today.
As a company where 60% of our team are Research and Development Engineers, we take special pride in continuous education. It’s what helps us stay on top of the game, both as individuals and as a company. Besides first-class mentorships, worldwide conferences, and relevant workshops, we heavily rely on books to keep our minds in shape. Our company library currently has more than 60 book titles and 4 periodical subscriptions, and it keeps growing.
If you want to stay up to date with the dynamic field of computer vision, we strongly advise you to read a lot. Be it books or online publications, each nugget of information may come handy at some point in your career.
Finally, if you’d like to continue your career in the field of computer vision as a part of the community that encourages learning and professional growth, check out our open positions.