You can deploy this model and start earning money today!

nihatxp/
clash-of-clans-base-segmentation

Name: clash-of-clans-base-segmentation
Availability: InStock

Views

No views yet

Clash of Clans - Base Area Segmentation (YOLOv8)

This is a custom YOLOv8-seg model trained specifically to detect the "Base Area" (the non-deployable village zone) in Clash of Clans screenshots.

By masking the actual village area, this model can be used by AI agents, automation tools, or game logic scripts to understand where troops cannot be deployed, perfectly mapping the playable boundaries of the game.

🔥 Validation Batch Predictions

The following image showcases the model's high-precision instance segmentation on a validation batch during training. The overlaid colored masks represent the exact "Base Area" identified by the AI.

📊 Technical Performance & Metrics

We provide comprehensive training metrics to demonstrate the robustness and accuracy of this model. It achieves a remarkable 98.5% mAP50 for the BaseArea class on validation data.

Training Results (Loss & mAP)

Mask F1-Confidence Curve

Normalized Confusion Matrix

💻 How to use

You can easily use this model with the official Ultralytics YOLO package:

pip install ultralytics opencv-python

python

import cv2
from ultralytics import YOLO

# Load the model directly from Hugging Face
model = YOLO("hf://nihatxp/clash-of-clans-base-segmentation/coc_deployable_seg.pt")

# Perform inference on your CoC screenshot
results = model.predict(source="your_village_screenshot.jpg", conf=0.25)

# Show the results
results[0].show()

🎯 Intended Use

This model is intended for educational purposes, computer vision research, and building non-malicious game assistants.