Koos du Preez
Koos du Preez
I am trying to run chirp on Python3 to help test the drivers, but seems the installer doesnt work and is looking for python package `wxPython` which i cant install...
> Ah - good point. sorry. I am running Ubuntu 20.04
> It would help to know what system you're running, but it works fine on python 3.8 without anything specific. See this page: https://chirp.danplanet.com/projects/chirp/wiki/DevelopersPython3Environment Seems there is no branch `py3`...
how big width/height in pixels are the items you are trying to detect? and how big resolution wise is your source image?
How many pixels width and height is 'not big' ?
if the network resizes 2048 down to 1024, you will get effective size of 15-30 pixels per object, which should be plenty to detect it.. But with a 1024x1024 network...
@chi8411 - Yes, you can cut the 2048x2048 image into 16 images of 512x512. You can use a sliding window of 512x512 with a stride of 480pixels or somethings similar,...
@wahid18benz, the anchor boxes should match the shape of your objects for better alignment and shape of predicted boxes. For example if you use a pre-defined network and its anchor...
@chi8411 - it is not part of yolo, i wrote it myself and yes, its processed into tiles of 512x512 (or whatever your network input size is) and then inference....
Gotcha. In this case, I would suggest you compute anchor boxes for this particular application.