OpenCV in C++ is great for doing computer vision and image processing work in a C++ environment. Now there is a Jupyter Kernel called Xeus for C++, which enables faster development cycles.

This repo contains a small example notebook showing how to use OpenCV in Xeus. It shows how to get the includes to work, and contains a small shim for displaying OpenCV images directly in the browser. It demonstrates a nice way to give small inline matrices.

To try it out: follow the Xeus installation directions, then run the notebook in this repository. It does a fairly simple task: loads an image, cuts the saturation by half, and displays the image. Interactively, you can try out and immediately see the effect of all sorts of changes, for example:

• Try changing the factor by which the saturation is changed from 2 to something else
• Try messing with the hue or value instead

It's not quite as convenient as the equivalent thing in python, and the technology is newer. But if you are using C++ anyway, your development iteration time may be sped up quite a bit.