--- license: apache-2.0 library_name: PaddleOCR language: - en - zh pipeline_tag: image-to-text tags: - OCR - PaddlePaddle - PaddleOCR - layout_detection --- # PicoDet-S_layout_17cls ## Introduction A high-efficiency layout area localization model trained on a self-built dataset of Chinese and English papers, magazines, and research reports using PicoDet-S. 17-Class Area Detection Model, including 17 common layout categories: Paragraph Title, Image, Text, Number, Abstract, Content, Figure Caption, Formula, Table, Table Caption, References, Document Title, Footnote, Header, Algorithm, Footer, and Seal. The key metrics are as follow: | Model| mAP(0.5) (%) | | --- | --- | |PicoDet-S_layout_17cls | 87.4 | **Note**: Paddleocr's self built layout area detection data set contains 892 common document type images such as Chinese and English papers, magazines and research papers. ## Quick Start ### Installation 1. PaddlePaddle Please refer to the following commands to install PaddlePaddle using pip: ```bash # for CUDA11.8 python -m pip install paddlepaddle-gpu==3.0.0 -i https://www.paddlepaddle.org.cn/packages/stable/cu118/ # for CUDA12.6 python -m pip install paddlepaddle-gpu==3.0.0 -i https://www.paddlepaddle.org.cn/packages/stable/cu126/ # for CPU python -m pip install paddlepaddle==3.0.0 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/ ``` For details about PaddlePaddle installation, please refer to the [PaddlePaddle official website](https://www.paddlepaddle.org.cn/en/install/quick). 2. PaddleOCR Install the latest version of the PaddleOCR inference package from PyPI: ```bash python -m pip install paddleocr ``` ### Model Usage You can quickly experience the functionality with a single command: ```bash paddleocr layout_detection \ --model_name PicoDet-S_layout_17cls \ -i https://cdn-uploads.huggingface.co/production/uploads/63d7b8ee07cd1aa3c49a2026/N5C68HPVAI-xQAWTxpbA6.jpeg ``` You can also integrate the model inference of the layout detection module into your project. Before running the following code, please download the sample image to your local machine. ```python from paddleocr import LayoutDetection model = LayoutDetection(model_name="PicoDet-S_layout_17cls") output = model.predict("N5C68HPVAI-xQAWTxpbA6.jpeg", batch_size=1, layout_nms=True) for res in output: res.print() res.save_to_img(save_path="./output/") res.save_to_json(save_path="./output/res.json") ``` After running, the obtained result is as follows: ```json {'res': {'input_path': '/root/.paddlex/predict_input/N5C68HPVAI-xQAWTxpbA6.jpeg', 'page_index': None, 'boxes': [{'cls_id': 2, 'label': 'text', 'score': 0.9770552515983582, 'coordinate': [35.47857, 350.32135, 359.99146, 607.66266]}, {'cls_id': 2, 'label': 'text', 'score': 0.9646613597869873, 'coordinate': [387.4421, 736.655, 712.7056, 850.04584]}, {'cls_id': 2, 'label': 'text', 'score': 0.9596860408782959, 'coordinate': [386.36847, 491.56995, 712.53467, 700.74225]}, {'cls_id': 2, 'label': 'text', 'score': 0.958072304725647, 'coordinate': [36.283585, 648.58374, 360.08328, 849.8185]}, {'cls_id': 8, 'label': 'table', 'score': 0.9491577744483948, 'coordinate': [62.92192, 104.4335, 330.56216, 299.23947]}, {'cls_id': 8, 'label': 'table', 'score': 0.9144826531410217, 'coordinate': [424.8911, 104.26025, 668.0119, 312.7304]}, {'cls_id': 0, 'label': 'paragraph_title', 'score': 0.8392149806022644, 'coordinate': [35.642235, 332.62488, 144.75916, 345.07657]}, {'cls_id': 0, 'label': 'paragraph_title', 'score': 0.796416699886322, 'coordinate': [389.69257, 717.66345, 526.9109, 729.3157]}, {'cls_id': 2, 'label': 'text', 'score': 0.6687009930610657, 'coordinate': [391.08237, 348.05475, 713.2955, 460.1097]}, {'cls_id': 2, 'label': 'text', 'score': 0.6419706344604492, 'coordinate': [35.950676, 21.344364, 361.3897, 79.71692]}, {'cls_id': 2, 'label': 'text', 'score': 0.5499911308288574, 'coordinate': [386.94125, 19.868908, 713.5697, 75.554535]}]}} ``` The visualized image is as follows: ![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/63d7b8ee07cd1aa3c49a2026/ObUtZYSYK9h3UHP4_fWQU.jpeg) For details about usage command and descriptions of parameters, please refer to the [Document](https://paddlepaddle.github.io/PaddleOCR/latest/en/version3.x/module_usage/layout_detection.html#iii-quick-integration). ### Pipeline Usage The ability of a single model is limited. But the pipeline consists of several models can provide more capacity to resolve difficult problems in real-world scenarios. #### PP-TableMagic (table_recognition_v2) The General Table Recognition v2 pipeline (PP-TableMagic) is designed to tackle table recognition tasks, identifying tables in images and outputting them in HTML format. PP-TableMagic includes the following 8 modules: * Table Structure Recognition Module * Table Classification Module * Table Cell Detection Module * Text Detection Module * Text Recognition Module * Layout Region Detection Module (optional) * Document Image Orientation Classification Module (optional) * Text Image Unwarping Module (optional) You can quickly experience the PP-TableMagic pipeline with a single command. ```bash paddleocr table_recognition_v2 -i https://cdn-uploads.huggingface.co/production/uploads/63d7b8ee07cd1aa3c49a2026/tuY1zoUdZsL6-9yGG0MpU.jpeg \ --layout_detection_model_name PicoDet-S_layout_17cls \ --use_doc_orientation_classify False \ --use_doc_unwarping False \ --save_path ./output \ --device gpu:0 ``` If save_path is specified, the visualization results will be saved under `save_path`. The command-line method is for quick experience. For project integration, also only a few codes are needed as well: ```python from paddleocr import TableRecognitionPipelineV2 pipeline = TableRecognitionPipelineV2( layout_detection_model_name=PicoDet-S_layout_17cls, use_doc_orientation_classify=False, # Use use_doc_orientation_classify to enable/disable document orientation classification model use_doc_unwarping=False, # Use use_doc_unwarping to enable/disable document unwarping module device="gpu:0", # Use device to specify GPU for model inference ) output = pipeline.predict("tuY1zoUdZsL6-9yGG0MpU.jpeg") for res in output: res.print() ## Print the predicted structured output res.save_to_img("./output/") res.save_to_xlsx("./output/") res.save_to_html("./output/") res.save_to_json("./output/") ``` The default model used in pipeline is `PP-DocLayout-L`, so it is needed that specifing to `PicoDet-S_layout_17cls` by argument `layout_detection_model_name`. And you can also use the local model file by argument `layout_detection_model_dir`. For details about usage command and descriptions of parameters, please refer to the [Document](https://paddlepaddle.github.io/PaddleOCR/main/en/version3.x/pipeline_usage/table_recognition_v2.html#2-quick-start). ## Links [PaddleOCR Repo](https://github.com/paddlepaddle/paddleocr) [PaddleOCR Documentation](https://paddlepaddle.github.io/PaddleOCR/latest/en/index.html)