Programming Computer Vision
with Python Jan Erik Solem

This is the website for my upcoming computer vision book. The book is meant as an entry point to hands-on computer vision for students, researchers and enthusiasts using the Python programming language.

Early draft

As I'm working to finish the book, I will post updated draft versions on this page. I encourage people to read the text, try the examples and send me comments.

The latest draft is February 15, 2012

Code and data

The code from the book is available here. Send me bugs and comments, it will help improve the book. As the book gets closer to completion, I'll take this code and make it into a proper project on github.

Most of the data used in the book is available from links in the text. All the other images and data sets can be downloaded here. (for credits, see the text for now)

Contact

Send your comments, thoughts, bugs etc by email.


Updates

[Feb 15] General clean up. Updated segmentation chapter. Reduced file size for easier download. Many small code fixes and clarifications. Klas J gave fantastic feedback.

[Jan 29] Updated OpenCV chapter to the new Python interface.

[Jan 15] Wrap up of chapter 5.

[Dec 13] Added section on stereo. Minor improvements and morphology to chapter 1.

[Nov 6] Last half of chapter 4 added.

[Oct 16] Wrap up of chapter 8. Updated the dataset. Good comments from Jens R and Neil Y, thanks.

[Sep 25] Added sections 8.2 and 8.3 to the classification chapter. Minor updates to the Harris corners section.
Thanks again for all comments, small and large. All reported typos and errors should have been addressed.

[Sep 18] General code cleanup of chapters 1-6. New Hierarchical clustering implementation. More on classification.
Thanks to all who submitted comments so far. Most of them have been incorporated in this version. Hakan Ardo deserves huge credit for his detailed comments. He contributed with lots of code cleanup in the first chapters and a complete refactor of hcluster.py.