Overwatch Aimbot powered by Deep Learning Object Detection models
ovWatch uses the YOLOv5 model for object detection. Currently, enemies are detected under the "Person" class (id:1) Git Repo: https://github.com/ultralytics/yolov5
The various libraries are used to move the cursor:
- pyautogui
- ctypes.windll
- Barebones image detection with pretrained YOLOv5
- Barebones aimbot functionality (Mouse movement to detected object)
- Support smoother enemy tracking
- Support auto screen rotation to detect more enemies
- Code out logic to determine when to stop firing (left click)
- Add toggle switch for aimbot
- Train model on Overwatch images to make it more accurate
- Support multiple attack types (Continuous hitscan, Burst hitscan, One-tap hitscan, Melee