YOLOv5 Kunshujo
YOLOv5 model fine-tuned for detecting layout elements in kunshujo (群集帖), a type of Japanese historical document.
Model Details
- Architecture: YOLOv5
- Framework: PyTorch
- Task: Object Detection
- Dataset: Kunshujo Layout Dataset
Usage
import torch
from huggingface_hub import hf_hub_download
model_path = hf_hub_download(repo_id="nakamura196/yolov5-kunshujo", filename="best.pt")
model = torch.hub.load("ultralytics/yolov5", "custom", path=model_path, source="local")
results = model("your_image.jpg")
results.show()
Demo
Try the interactive demo on Hugging Face Spaces: nakamura196/yolov5-kunshujo
Examples
The model is trained to detect layout elements in historical Japanese documents such as 張交帖 (National Diet Library, Japan).