How to Determine Image Dimensions Efficiently?

[b-simjoo]

I'm looking for a better way to determine image dimensions before initializing shared memory. The current approach—connecting to the camera with OpenCV, retrieving the image size, and then setting up shared memory—isn't optimal.

If there's a more efficient solution, I couldn't find it in the tutorials. Any guidance or suggestions would be greatly appreciated!

[elsampsa]

A quick comment: that god'damn "readthedocs.io" - not sure how it gets generated (!). Please use always valkka.org.
Will comment on your actual issue in a while..

[elsampsa]

In this lesson: https://elsampsa.github.io/valkka-examples/_build/html/lesson_4.html
what you actually do, is:
Define the filterchain - while doing that, you also define the expected image dimensions. This before starting any threads or multiprocesss - so you do know them beforehand any connection to the camera.