Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,56 @@
|
|
| 1 |
-
|
| 2 |
-
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
<!-- ABOUT THE PROJECT -->
|
| 2 |
+
## About 📋
|
| 3 |
+
|
| 4 |
+
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.
|
| 5 |
+
|
| 6 |
+
<p align="center">
|
| 7 |
+
<img src="https://github.com/moured/YOLOv10-Document-Layout-Analysis/raw/main/images/samples.gif" height="320"/>
|
| 8 |
+
</p>
|
| 9 |
+
|
| 10 |
+
## Results 📊
|
| 11 |
+
| Model | mAP50 | mAP50-95 | Model Weights |
|
| 12 |
+
|---------|-------|----------|---------------|
|
| 13 |
+
| YOLOv10-x | 0.924 | 0.740 | [Download](https://github.com/moured/YOLOv10-Document-Layout-Analysis/releases/download/doclaynet_weights/yolov10x_best.pt) |
|
| 14 |
+
| YOLOv10-b | 0.922 | 0.732 | [Download](https://github.com/moured/YOLOv10-Document-Layout-Analysis/releases/download/doclaynet_weights/yolov10b_best.pt) |
|
| 15 |
+
| YOLOv10-l | 0.921 | 0.732 | [Download](https://github.com/moured/YOLOv10-Document-Layout-Analysis/releases/download/doclaynet_weights/yolov10l_best.pt) |
|
| 16 |
+
| YOLOv10-m | 0.917 | 0.737 | [Download](https://github.com/moured/YOLOv10-Document-Layout-Analysis/releases/download/doclaynet_weights/yolov10m_best.pt) |
|
| 17 |
+
| YOLOv10-s | 0.905 | 0.713 | [Download](https://github.com/moured/YOLOv10-Document-Layout-Analysis/releases/download/doclaynet_weights/yolov10s_best.pt) |
|
| 18 |
+
| YOLOv10-n | 0.892 | 0.685 | [Download](https://github.com/moured/YOLOv10-Document-Layout-Analysis/releases/download/doclaynet_weights/yolov10n_best.pt) |
|
| 19 |
+
|
| 20 |
+
### Installation 💻
|
| 21 |
+
```
|
| 22 |
+
conda create -n yolov10 python=3.9
|
| 23 |
+
conda activate yolov10
|
| 24 |
+
git clone https://github.com/THU-MIG/yolov10.git
|
| 25 |
+
cd yolov10
|
| 26 |
+
pip install -r requirements.txt
|
| 27 |
+
pip install -e .
|
| 28 |
+
```
|
| 29 |
+
|
| 30 |
+
## References 📝
|
| 31 |
+
|
| 32 |
+
1. YOLOv10
|
| 33 |
+
```
|
| 34 |
+
BibTeX
|
| 35 |
+
@article{wang2024yolov10,
|
| 36 |
+
title={YOLOv10: Real-Time End-to-End Object Detection},
|
| 37 |
+
author={Wang, Ao and Chen, Hui and Liu, Lihao and Chen, Kai and Lin, Zijia and Han, Jungong and Ding, Guiguang},
|
| 38 |
+
journal={arXiv preprint arXiv:2405.14458},
|
| 39 |
+
year={2024}
|
| 40 |
+
}
|
| 41 |
+
```
|
| 42 |
+
|
| 43 |
+
|
| 44 |
+
2. DocLayNet
|
| 45 |
+
```
|
| 46 |
+
@article{doclaynet2022,
|
| 47 |
+
title = {DocLayNet: A Large Human-Annotated Dataset for Document-Layout Analysis},
|
| 48 |
+
doi = {10.1145/3534678.353904},
|
| 49 |
+
url = {https://arxiv.org/abs/2206.01062},
|
| 50 |
+
author = {Pfitzmann, Birgit and Auer, Christoph and Dolfi, Michele and Nassar, Ahmed S and Staar, Peter W J},
|
| 51 |
+
year = {2022}
|
| 52 |
+
}
|
| 53 |
+
```
|
| 54 |
+
|
| 55 |
+
## Contact
|
| 56 |
+
LinkedIn: [https://www.linkedin.com/in/omar-moured/](https://www.linkedin.com/in/omar-moured/)
|