This repository provides checkpoints for YOLOv8, RT-DETR, and YOLOv10 models, each fine-tuned on the VisDrone dataset. These models are optimized specifically for detecting UAV (Unmanned Aerial Vehicle) images.

├── LICENSE            <- Open-source license if one is chosen
├── README.md          <- The top-level README for developers using this project.
├── demo
│
├── docs               <- A default mkdocs project; see mkdocs.org for details
│
├── uav_models         <- Trained and serialized models, model predictions, or model summaries
│
│             
├── reports            <- Generated analysis as HTML, PDF, LaTeX, etc.
│   └── figures        <- Generated graphics and figures to be used in 
│
├── requirements.txt   <- The requirements file for reproducing the analysis environment, e.g.
                          generated with `pip freeze > requirements.txt`

• YOLOv8 is a state-of-the-art object detection model that offers high accuracy and speed. It is used in this project for vehicle detection tasks.

• RT-DETR (Real-Time Detection Transformer) is another powerful model utilized for object detection. It leverages transformer architecture to enhance detection capabilities, providing robust results in real-time scenarios.

• YOLOv10 is an advanced version of the YOLO series models, offering improved performance and accuracy. It is included in this project to explore its effectiveness in vehicle detection.

1. Clone the repository:

   git clone https://github.com/NhiNguyen34/uav-detection.git
   cd uav-detection

2. Create and activate the conda environment:

   On Linux/macOS:

   # Create a new conda environment
   conda create --name visdrone-det python=3.10 -y
   
   # Activate the environment
   source activate visdrone-det
   
   # Install dependencies
   pip install -r requirements.txt
   
   echo "Setup complete. Please place your model checkpoints in the 'checkpoints' folder and your demo images in the 'demo' folder."

   On Windows:

   :: Create a new conda environment
   conda create --name visdrone-det python=3.10 -y
   
   :: Activate the environment
   conda activate visdrone-det
   
   :: Install dependencies
   pip install -r requirements.txt
   
   echo Setup complete. Please place your model checkpoints in the 'checkpoints' folder and your demo images in the 'demo' folder.

3. Save the model checkpoint: Download your model checkpoint and save it in the checkpoints folder. Ensure the file is named appropriately, e.g., 4epoch-rtdetr-best.pt.

4. Run the model:

   python run.py --model_path checkpoints/your_model_checkpoint.pt --image_path demo/your_image.jpg --model_detect "your_model" 

• YOLOv8 Documentation: YOLOv8 GitHub
• RT-DETR Paper: Real-Time Detection Transformer
• YOLOv10 Documentation: YOLOv10 GitHub