An auto-controller for Minecraft utilizing OpenCV & OpenCV's Deep Nerual Network Module. Features modes like, avoidance, running, harvesting, mining, and guarding. Detections utilized by trained YoloV5s models converted to ONNX.
The program provides some neat features such as...
Safe Walk
Fighting
Harvesting
Mining
Sentry
Auto Feeder
Auto Unstuck
Four Detectable Enemies
- Creepr
- Zombie
- Spider
- Skeleton
Two Detectable Trees
- Oak
- Birch
Seven Detectable Ores
- Coal
- Diamond
- Redstone
- Gold
- Iron
- Lapis
- Redstone
- Poor detection at close and at longer ranges
- Easily looses track of enemies & other detections
- Cannot fight players
- Cannot fight most enemies
- Cannot harvest most trees
- Realistically can fight 1-2 enemies safely
- Visaul Studio 2019
- Cuda 11.2
- Cudnn 8.8
- OpenCV 4.5.5
- i7 10700k
- RTX 3070
- 32 GB DDR4 Ram @ 3200MHz
- M.2 SSD