This project provides a backend API for detecting military aircraft in images. It leverages a deep learning model built with PyTorch, offering an open-weight architecture for transparency and further development. The model uses SqueezeNet as a backbone and RetinaNet-based heads for accurate object detection. The API itself is constructed using FastAPI, ensuring efficient and robust performance.
- Military Aircraft Detection: Identifies and localizes military aircraft within images.
- Open-Weight Model: The underlying PyTorch model and weights are publicly available, allowing for inspection, modification, and fine-tuning.
- FastAPI Backend: Provides a high-performance and easy-to-use API for interacting with the model.
- PyTorch: Deep learning framework for building and training the detection model.
- FastAPI: Modern, fast (high-performance), web framework for building APIs with Python 3.7+ based on standard Python type hints.
- Python: Primary programming language.
For development run
source .venv/bin/activate
fastapi dev main.pyFor production run
source .venv/bin/activate
fastapi run main.pyThis project continues the competition for the master "Applied Artificial Intelligence" AAIV 2025-I:Object Location
This project utilizes several open-source resources and libraries. We would like to acknowledge the contributions of the PyTorch, SqueezeNet, RetinaNet, and FastAPI communities.