Me try to learn [ computer vision | object detection ] on Ragnarok game.
===
I'm so excited to learn how to make a model that can detect objects in a game! I want it to be able to spot monsters, item drops, and dangerous situations where a character's HP is almost 0%.
So I think I need to breakdown the tasks:
- Train a new dataset by taking a lot of in-game screenshots.
- Use labeling application to help labeling the images.
- Train the labeled data into a model.
- ...
- X11
- scrot => 1.11.1 (support WID)
- wmctrl
python -m venv venv
isolates project environment.source venv/bin/activate
activates project environment.make screenshot
starts screenshot on primary monitor (if you use multi monitors).