yasirfaizahmed/android_ui_detection_yolov8

Android UI Detection – YOLOv8

This YOLOv8 model is trained to detect various Android UI elements in app/game screenshots, such as buttons, cards, toolbars, text views, and more.

Trained using YOLOv8 Nano
Detects 21 Android UI classes
Ideal for UI automation, testing, and design analysis

---

Installation

pip install ultralytics

---

How to Load and Use the Model

from ultralytics import YOLO

# Load the model directly from Hugging Face
model = YOLO("yasirfaizahmed/android_ui_detection_yolov8")

# Run detection on an image
results = model("your_image.jpg")  # Replace with your actual image path

# Show results with bounding boxes
results[0].show()

---

Classes Detected

[
  'BackgroundImage', 'Bottom_Navigation', 'Card', 'CheckBox', 'Checkbox',
  'CheckedTextView', 'Drawer', 'EditText', 'Icon', 'Image', 'Map', 'Modal',
  'Multi_Tab', 'PageIndicator', 'Remember', 'Spinner', 'Switch', 'Text',
  'TextButton', 'Toolbar', 'UpperTaskBar'
]

---

Model Structure

• Trained with: yolov8n.pt base

• Format: YOLOv8 PyTorch

• Dataset: Custom Pascal VOC-style Android UI dataset

---

Training Configuration

• Recommended image size: 640×640

• Supports predict, val, export, and train pipelines from Ultralytics

• Use .predict(source="folder_or_image.jpg") for batch inference

---

[More Information Needed]