anyformat/doclayout-yolo-docstructbench

DocLayout-YOLO - Docstructbench

Document layout detection model. Paper: DocLayout-YOLO

Classes (10)

• title
• plain_text
• abandon
• figure
• figure_caption
• table
• table_caption
• table_footnote
• isolate_formula
• formula_caption

Usage

pip install anyformat-doclayout

from anyformat.doclayout import DocLayoutModel, download_converted

# Download weights from this repo
weights = download_converted("docstructbench")

# Run inference
model = DocLayoutModel(weights)
results = model.predict("document.png")

for det in results:
    print(f"{det['class_name']}: {det['confidence']:.2f}")