This project aims to apply computer vision in games to automate repetitive tasks. My main goal is to study the fields of computer vision and machine learning while gaming. It uses YoloV4 and OpenCV to identify "crops" in the game "Albion Online¨.
- General Info
- Technologies Used
- Features
- Screenshots
- Setup
- Usage
- Project Status
- Room for Improvement
- Acknowledgements
- Contact
- Provide general information about your project here. We are using a custom trained detection model to detect 'crops' on AlbionOnline, this can later be used to automate the farming labor of the game. We locate the game window based on its name (Linux only), then capture screenshots and send them to the dectetion model to detect and classify the objetcs (crops), after that we are planning to automate the task of collecting grown crops.
- What problem does it (intend to) solve? This project aims to apply computer vision in games to automate repetitive tasks.
- What is the purpose of your project? The purpose of this project is to apply computer vision and machine learning to automate certain games' tasks. Like colecting resources, crafting or even combat.
- OpenCV
- YoloV4
- PyAutoGUI
- Fastgrab API
- Xdo API
- Numpy
List the ready features here:
- Object Detection in Real Time
Project is: in progress / very early development stage
Room for improvement:
- Threads are out of sync, we could sync them and fix it schelude.
- It's taking screenshots way faster than the detection model can handle.
To do:
- Fix threads timing
- Add the automation to colect grown crops
- Add the automation to plant new chosen crops
- Add herbs to the system.
Give credit here.
- This project was inspired by the youtube channel Learn Code By Gaming.
- This project was based on this tutorial.
- Many thanks to Ben from channel Learn Code By Gaming, for inspiring me to start this new project and learn a new language and this incredible field of computer vision and machine learning.
Created by @flan6 - feel free to contact me!