How to use Riksarkivet/yolov9-lines-within-regions-1 with HTRflow:
# CLI usage # see docs: https://ai-riksarkivet.github.io/htrflow/latest/getting_started/quick_start.html htrflow pipeline <path/to/pipeline.yaml> <path/to/image>
# Python usage from htrflow.pipeline.pipeline import Pipeline from htrflow.pipeline.steps import Task from htrflow.models.framework.model import ModelClass pipeline = Pipeline( [ Task( ModelClass, {"model": "Riksarkivet/yolov9-lines-within-regions-1"}, {} ), ])
How to use Riksarkivet/yolov9-lines-within-regions-1 with ultralytics:
# Couldn't find a valid YOLO version tag. # Replace XX with the correct version. from ultralytics import YOLOvXX model = YOLOvXX.from_pretrained("Riksarkivet/yolov9-lines-within-regions-1") source = 'http://images.cocodataset.org/val2017/000000039769.jpg' model.predict(source=source, save=True)
It wont work on full pages but only on cropped text-regions.
Works absolutely fine on full pages. Your region model also regularly makes the full page the region.
· Sign up or log in to comment
