Downloads · 30 days
5
16% of all-time downloads
nethsith/sri-lankan-album-stamp-detector-yolo
sri-lankan-album-stamp-detector-yolo is a object detection model from nethsith. Use it when you need objects located in an image. It is set up for ultralytics. The card lists the license as mit.
This model is a YOLO-based object detection model trained to detect stamps and paper labels from Sri Lankan stamp album page images.
Downloads · 30 days
5
16% of all-time downloads
All-time downloads
31
Public
Repo size
6.1 MB
Likes
0
Public
Click a slice to open those files.
.pt5.5 MB · 90%
From the Hugging Face model README
This model is a YOLO-based object detection model trained to detect stamps and paper labels from Sri Lankan stamp album page images.
It is used as Model 1 in the Stamp AI project, where the goal is to detect individual stamps from full album pages and extract stamp crops for later matching.
0: stamp
1: paper_label
This model is intended for:
This model was trained using the Sri Lankan Album Stamp Detection Dataset:
https://huggingface.co/datasets/nethsith/sri-lankan-album-stamp-detection
Expected files in this model repository:
best.pt
dataset.yaml
README.md
results.png
confusion_matrix.png
BoxF1_curve.png
BoxPR_curve.png
Install Ultralytics:
pip install ultralytics
Run prediction:
from ultralytics import YOLO
model = YOLO("best.pt")
results = model.predict(
source="path/to/album_page.jpg",
conf=0.36,
iou=0.20,
save=True,
save_crop=True,
classes=[0]
)
Use classes=[0] when you only want to crop detected stamps.
The model returns bounding boxes around detected objects. When save_crop=True is used, detected stamp crops are saved automatically by Ultralytics.
Detection quality may be affected by:
The model is intended to support stamp crop extraction, not final stamp identity matching.
This model is released under the MIT License.
Please cite or credit this model as:
Sri Lankan Album Stamp Detector YOLO by Nethsith Gunaweera.