| <!-- ABOUT THE PROJECT --> | |
| ## About 📋 | |
| The models were fine-tuned using 4xA100 GPUs on the Doclaynet-base dataset, which consists of 6910 training images, 648 validation images, and 499 test images. | |
| <p align="center"> | |
| <img src="https://github.com/moured/YOLOv10-Document-Layout-Analysis/raw/main/images/samples.gif" height="320"/> | |
| </p> | |
| ## Results 📊 | |
| | Model | mAP50 | mAP50-95 | Model Weights | | |
| |---------|-------|----------|---------------| | |
| | YOLOv10-x | 0.924 | 0.740 | [Download](https://github.com/moured/YOLOv10-Document-Layout-Analysis/releases/download/doclaynet_weights/yolov10x_best.pt) | | |
| | YOLOv10-b | 0.922 | 0.732 | [Download](https://github.com/moured/YOLOv10-Document-Layout-Analysis/releases/download/doclaynet_weights/yolov10b_best.pt) | | |
| | YOLOv10-l | 0.921 | 0.732 | [Download](https://github.com/moured/YOLOv10-Document-Layout-Analysis/releases/download/doclaynet_weights/yolov10l_best.pt) | | |
| | YOLOv10-m | 0.917 | 0.737 | [Download](https://github.com/moured/YOLOv10-Document-Layout-Analysis/releases/download/doclaynet_weights/yolov10m_best.pt) | | |
| | YOLOv10-s | 0.905 | 0.713 | [Download](https://github.com/moured/YOLOv10-Document-Layout-Analysis/releases/download/doclaynet_weights/yolov10s_best.pt) | | |
| | YOLOv10-n | 0.892 | 0.685 | [Download](https://github.com/moured/YOLOv10-Document-Layout-Analysis/releases/download/doclaynet_weights/yolov10n_best.pt) | | |
| ### Installation 💻 | |
| ``` | |
| conda create -n yolov10 python=3.9 | |
| conda activate yolov10 | |
| git clone https://github.com/THU-MIG/yolov10.git | |
| cd yolov10 | |
| pip install -r requirements.txt | |
| pip install -e . | |
| ``` | |
| ## References 📝 | |
| 1. YOLOv10 | |
| ``` | |
| BibTeX | |
| @article{wang2024yolov10, | |
| title={YOLOv10: Real-Time End-to-End Object Detection}, | |
| author={Wang, Ao and Chen, Hui and Liu, Lihao and Chen, Kai and Lin, Zijia and Han, Jungong and Ding, Guiguang}, | |
| journal={arXiv preprint arXiv:2405.14458}, | |
| year={2024} | |
| } | |
| ``` | |
| 2. DocLayNet | |
| ``` | |
| @article{doclaynet2022, | |
| title = {DocLayNet: A Large Human-Annotated Dataset for Document-Layout Analysis}, | |
| doi = {10.1145/3534678.353904}, | |
| url = {https://arxiv.org/abs/2206.01062}, | |
| author = {Pfitzmann, Birgit and Auer, Christoph and Dolfi, Michele and Nassar, Ahmed S and Staar, Peter W J}, | |
| year = {2022} | |
| } | |
| ``` | |
| ## Contact | |
| LinkedIn: [https://www.linkedin.com/in/omar-moured/](https://www.linkedin.com/in/omar-moured/) |