RF-DETR Stamp/Signature/Checkbox Detector

A fine-tuned RF-DETR model for detecting stamps, signatures, and checkboxes in document images.

Model Details

Base Model: RF-DETR Base (DINOv2 backbone)
Training: 30 epochs on 8x NVIDIA H100 80GB
Dataset: ~9,400 images (merged from multiple sources)
License: Apache 2.0

Classes

ID	Class
0	stamp
1	signature
2	checkbox_checked
3	checkbox_unchecked

Performance

Class	mAP@50	mAP@50:95	Precision	Recall
stamp	93.2%	70.5%	96.3%	81%
signature	95.4%	73.9%	96.6%	81%
checkbox_checked	91.8%	56.5%	90.7%	81%
checkbox_unchecked	65.7%	34.2%	43.4%	81%
overall	86.5%	58.8%	81.7%	81%

Usage

from rfdetr import RFDETRBase
from PIL import Image

# Load model
model = RFDETRBase()
model.load("path/to/checkpoint_best_ema.pth")

# Run inference
image = Image.open("document.png")
detections = model.predict(image, threshold=0.5)

for det in detections:
    print(f"Class: {det['class']}, Confidence: {det['confidence']:.2f}, Box: {det['bbox']}")

Training Details

Batch size: 32 per GPU (256 effective)
Learning rate: 4e-4
Optimizer: AdamW
Resolution: 560px

Limitations

checkbox_unchecked detection is weaker (65.7% mAP) due to low visual distinctiveness
Best suited for stamps and signatures in document images

Citation

@misc{bluecopa-stamp-detector,
  title={RF-DETR Stamp/Signature Detector},
  author={BlueCopa},
  year={2024},
  publisher={HuggingFace}
}

Downloads last month: 3