Indoor vs Outdoor Classifier
A binary image classifier that determines whether a property image is indoor or outdoor with 96% accuracy.
Model Details
- Model Type: MobileNetV2-based binary classifier
- Classes: Indoor, Outdoor
- Accuracy: 96% on validation set
- Input Size: 160x160 RGB images
Training Data
The model was trained on 2,000 curated Airbnb property images, split evenly between indoor and outdoor scenes. The dataset was manually verified to ensure high-quality training examples.
Use Cases
- Real estate listing automation
- Property image organization
- Virtual tour preparation
- Interior design vs. architecture applications
License
This model is released under MIT License.
Code Implementation
Example usage code has been provided in example_usage.py under model files.
