HMPNet：A Feature Aggregation Architecture for Maritime Object Detection from a Shipborne Perspective

🎉 Accepted at ICME2025! 🎉

View Paper

---

🚀 Introduction

We are thrilled to announce that our paper HMPNet: A Feature Aggregation Architecture for Maritime Object Detection from a Shipborne Perspective has been accepted for presentation at ICME2025! This marks a significant milestone in our journey, reflecting countless hours of research, development, and unwavering perseverance.

"Hard work and dedication always pay off."

---

🔧 How to Use

Follow the instructions below to set up and experiment with HMPNet.

---

🖥️ Environment

• GPU: NVIDIA RTX 4090
• Python: 3.10
• PyTorch: 2.0.2

---

📚 Table of Contents

• ⚙️ Installation
• 📦 Dataset Preparation
• 🚀 Training & Evaluation
• 📄 Citation
• 🤝 Acknowledgement
• 📜 License

🛠️ Installation

1. Clone the Repository

git clone https://github.com/yourusername/HMPNet.git
cd HMPNet

2. Create and Activate a Python Virtual Environment

conda create -n HMPNet python=3.10
conda activate HMPNet

3. Install the Required Dependencies

pip install -r requirements.txt

---

📁 Dataset Preparation

1. Download the maritime object detection dataset (or use your own dataset).
2. Organize the dataset in YOLO format:

├── dataset/
    ├── train/
        ├── images/
        ├── labels/
    ├── val/
        ├── images/
        ├── labels/
    ├── test/
        ├── images/
        ├── labels/

3. Update the dataset path in the configuration file (e.g., data.yaml).

---

🎓 Training and Evaluation

Training from Scratch

python train.py

Validation

python val.py

Testing

python val.py split='test'

---

🌟 Acknowledgements

We sincerely thank the following projects for their inspiring and foundational contributions:

• ultralytics
• PaddleDetection
• ParameterNet
• DEA-Net
• RTDETR

Your wonderful code and insights have greatly influenced our work.

---

📝 Citation

If you find this repository useful, please consider citing our paper:

@article{yourpaper2025,
  title={HMPNet: A Feature Aggregation Architecture for Maritime Object Detection},
  author={Your Name and Collaborators},
  journal={ICME2025 Proceedings},
  year={2025}
}

---

📜 License

This project is licensed under the GPL-3.0 License. See the LICENSE file for details.

---

Committed to advancing maritime object detection research one step at a time.