Rating:
(8 reviews)
Author: Dr Simon J. D. Prince
ISBN : 1107011795
New from $81.00
Format: PDF
Free download Free Computer Vision: Models, Learning, and Inference Hardcover from 4shared, mediafire, hotfile, and mirror link
Review
"Computer vision and machine learning have gotten married and this book is their child. It gives the machine learning fundamentals you need to participate in current computer vision research. It's really a beautiful book, showing everything clearly and intuitively. I had lots of 'aha!' moments as I read through the book. This is an important book for computer vision researchers and students, and I look forward to teaching from it."
William T. Freeman, Massachusetts Institute of Technology
"With clarity and depth, this book introduces the mathematical foundations of probabilistic models for computer vision, all with well-motivated, concrete examples and applications. Most modern computer vision texts focus on visual tasks; Prince's beautiful new book is natural complement, focusing squarely on fundamental techniques, emphasizing models and associated methods for learning and inference. I think every serious student and researcher will find this book valuable. I've been using draft chapters of this remarkable book in my vision and learning courses for more than two years. It will remain a staple of mine for years to come."
David J. Fleet, University of Toronto
"This book addresses the fundamentals of how we make progress in this challenging and exciting field. I look forward to many decades with [this book] on my shelf, or indeed, I suspect, open on my desktop."
from the Foreword by Andrew Fitzgibbon
"Prince's magnum opus provides a fully probabilistic framework for understanding modern computer vision. With straightforward descriptions, insightful figures, example applications, exercises, background mathematics, and pseudocode, this book is self-contained and has all that is needed to explore this fascinating discipline."
Roberto Cipolla, University of Cambridge
"The author's goal, as stated in the preface, is to provide a book that focuses on the models involved, and I think the book has succeeded in doing that. I learned quite a bit and would recommend this text highly to the motivated, mathematically mature reader."
Jeffrey Putnam, Computing Reviews
Book Description
With minimal prerequisites, the book starts from the basics of probability and model fitting and works up to real examples that the reader can implement and modify to build useful vision systems. Primarily meant for advanced undergraduate and graduate students, the detailed methodological presentation will also be useful for practitioners of computer vision.
See all Editorial Reviews
Direct download links available for Free Computer Vision: Models, Learning, and Inference Hardcover
- Hardcover: 598 pages
- Publisher: Cambridge University Press (June 18, 2012)
- Language: English
- ISBN-10: 1107011795
- ISBN-13: 978-1107011793
- Product Dimensions: 1.3 x 6.9 x 10 inches
- Shipping Weight: 3.1 pounds (View shipping rates and policies)
Free Computer Vision: Models, Learning, and Inference
I teach the Machine Vision class at UCL from this textbook (for advanced undergrads + grad students). It's the same class Simon Prince used to teach, so we cover the whole book (ok, skipping a few bits and one whole chapter) in 11 weeks of lectures. The two main reasons I like it are 1) its unified explanation of all the major topics, and 2) the extra materials for students and teachers (free online):
1) Everything is explained in terms of (essentially) the same probabilistic models. That probably doesn't sound seriously exciting, but imagine the alternative of having to learn all the complicated math for doing object recognition, camera pose estimation, tracking, pose regression, shape modeling etc, but each one using ITS OWN notation and completely different "slices" of applied machine learning! It was hard to learn, and very hard to teach. Here, almost everything is consistent (even Structure from Motion is somehow made to fit the same notation). So if you can survive Chapters 2-4 (spread gently over ~40 pages), you'll likely absorb the rest without the usual agony.
2) On the book's website, Prince has built a collection of slides (pretty plain, but good), and an AMAZING (still evolving?) 75-page booklet of algorithms. While the textbook is accurate, there's normally quite some head-scratching to turn the equations into code. You obviously still have to write the code yourself, but now you have a recipe! It's clear the book would be unreadable if each algorithm's details had been included in the main text, so this seems like an ok compromise. This really could be the next "Numerical Recipes in C," but for vision :) There are interesting links to other people's data and code online too, and solutions to some of the problem sets.