Detection

Previously, I built an inference app using Hugging Face Spaces and a YOLOv5 model (trained on the NDL-DocL dataset): This time, I modified the app above to add JSON output, as shown in the following diff: https://huggingface.co/spaces/nakamura196/yolov5-ndl-layout/commit/4d48b95ce080edd28d68fba2b5b33cc17b9b9ecb#d2h-120906 This enables processing using the returned results, as shown in the following notebook: https://github.com/nakamura196/ndl_ocr/blob/main/GradioのAPIを用いた物体検出例.ipynb There may be better approaches, but I hope this serves as a useful reference.

Overview I built a layout extraction model using the NDL-DocL dataset and YOLOv5. https://github.com/ndl-lab/layout-dataset https://github.com/ultralytics/yolov5 You can try this model using the following notebook. https://colab.research.google.com/github/nakamura196/ndl_ocr/blob/main/NDL_DocLデータセットとYOLOv5を用いたレイアウト抽出モデル.ipynb This article is a record of the training process above. Creating the Dataset The NDL-DocL dataset in Pascal VOC format is converted to YOLO format. For this method, refer to the following article. In addition to the conversion from Pascal VOC format to COCO format, conversion from COCO format to YOLO format was added. ...