Codelab for implementing an image stitching pipeline with Sift, using Python, OpenCV and Numpy.

The codelab contains a guided track and a to be completed source code that gives an introduction on Sift and images stitching. Moreover there is also a jupiter notebook that follows the guided track step by step.

The lab comprises a theorical introduction to Sift and features matching. Then, few parts of the code are left to be completed or understood by the students.

The codelab requires the you have Python, Numpy and OpenCV with Python bindings correctly installed on your system.

If you were not able to set up a proper environment there is also a docker already configured.

1. Install docker (for both Linux, Mac and Windows): https://docs.docker.com/engine/installation/
2. Point your shell (terminal, power shell or whatever) in this directory and then type: docker build -t optimus_prime .
3. I suggest to follow this nice warm-up: https://docs.docker.com/, then in order to start the docker container type: docker run -d -p 8888:8888 -v /your/directory/with/this/files:/home optimus_prime
4. Open your browser on http://localhost:8888, select Hacking Sift.ipynb notebook and enjoy!

You are free to use this codelab for your seminars, for you courses or for your own education, but you have to give credits at the author for his work. The copyright of this work is owned by the author Alessandro Ferrari. The materials including the source codes and the latex sources are released with gpl-3.0 license. You should have received a copy of the GNU General Public License along with this tutorial. If not, see http://www.gnu.org/licenses/. The codelab tutorial HackingSift.pdf and the images in the img folder are released under Creative Commons Attribution-ShareAlike 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by-sa/4.0/.