YOLOv8 model to detect layout of 寛政重修諸家譜
Inference
Supported Labels
["h1", "h2", "head", "long", "other", "page", "sode", "text", "vol"]
How to use
pip install ultralyticsplus
from ultralyticsplus import YOLO, render_result
# load model
model = YOLO('best.pt')
# set image
image = "https://dl.ndl.go.jp/api/iiif/1879314/R0000039/full/640,640/0/default.jpg"
# perform inference
results = model.predict(image, verbose=False)
# observe results
render = render_result(model=model, image=image, result=results[0])
render.show()
- Downloads last month
- 7
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the HF Inference API does not support ultralytics models with pipeline type object-detection
Evaluation results
- mAP50-95(B)self-reported0.671
- mAP50(B)self-reported0.947