PP-StructureV3 Service Deployment & API Usage Example:

PaddleOCR open-source project GitHub address, the code here is completely consistent with the open-source model.

1. Introduction to the PP-StructureV3 Pipeline

PP-StructureV3 is an efficient and comprehensive document analysis solution capable of extracting structured information from various types of document images and PDF files. By integrating advanced technologies such as Optical Character Recognition (OCR), image processing, and deep learning, PP-StructureV3 can identify and extract elements like text blocks, headings, paragraphs, images, tables, formulas, and charts from documents. It transforms complex document contents into machine-readable data formats (such as Markdown and JSON), greatly improving the efficiency and accuracy of document data processing.

The PP-StructureV3 pipeline mainly consists of the following modules:

• Preprocessing Module: Includes document image orientation classification and text image correction, providing high-quality input for subsequent processing.
• Layout Region Detection Module: Precisely locates and differentiates various functional regions in the document through layout region detection.
• OCR Recognition Module: Utilizes the PP-OCRv5 model to efficiently recognize text content in detected text regions.
• Document Element Recognition Module (Optional): Supports table recognition, formula recognition, chart parsing, and stamp recognition, enabling extraction of rich structured semantic information from various complex documents.
• Reading Order Module: Restores the reading order and layout structure of multi-column documents, making the results more consistent with the logical structure of the original document.

PP-StructureV3 performs excellently across various document types, especially when processing documents with complex structures and multi-column layouts. The following diagram shows the overall workflow of the PP-StructureV3 document analysis pipeline:

2. API Description

This applies to scenarios where parsing tasks are created via API. Users must first apply for a Token. Note:

• Each user has a daily limit of 3000 pages per model for parsing. Further parsing requests will return a 429 error.
• There is no limit on the number of pages or file size for a single file, but if large files cause resource constraints, the service may time out and return a 503 error.

3. Service Call Example (python)

# Please make sure the requests library is installed
# pip install requests
import base64
import os
import requests

# Please visit https://aistudio.baidu.com/paddleocr/task to obtain the API_URL and TOKEN in the API call example.
API_URL = "<your url>"
TOKEN = "<access token>"

file_path = "<local file path>"

with open(file_path, "rb") as file:
    file_bytes = file.read()
    file_data = base64.b64encode(file_bytes).decode("ascii")

headers = {
    "Authorization": f"token {TOKEN}",
    "Content-Type": "application/json"
}

required_payload = {
    "file": file_data,
    "fileType": <file type>,  # For PDF documents, set `fileType` to 0; for images, set `fileType` to 1
}

optional_payload = {
    "useDocOrientationClassify": False,
    "useDocUnwarping": False,
    "useTextlineOrientation": False,
    "useChartRecognition": False,
}

payload = {**required_payload, **optional_payload}


response = requests.post(API_URL, json=payload, headers=headers)
print(response.status_code)
assert response.status_code == 200
result = response.json()["result"]

output_dir = "output"
os.makedirs(output_dir, exist_ok=True)

for i, res in enumerate(result["layoutParsingResults"]):
    md_filename = os.path.join(output_dir, f"doc_{i}.md")
    with open(md_filename, "w") as md_file:
        md_file.write(res["markdown"]["text"])
    print(f"Markdown document saved at {md_filename}")
    for img_path, img in res["markdown"]["images"].items():
        full_img_path = os.path.join(output_dir, img_path)
        os.makedirs(os.path.dirname(full_img_path), exist_ok=True)
        img_bytes = requests.get(img).content
        with open(full_img_path, "wb") as img_file:
            img_file.write(img_bytes)
        print(f"Image saved to: {full_img_path}")
    for img_name, img in res["outputImages"].items():
        img_response = requests.get(img)
        if img_response.status_code == 200:
            # Save image to local
            filename = os.path.join(output_dir, f"{img_name}_{i}.jpg")
            with open(filename, "wb") as f:
                f.write(img_response.content)
            print(f"Image saved to: {filename}")
        else:
            print(f"Failed to download image, status code: {img_response.status_code}")

Main operations provided by the service:

• The HTTP request method is POST.
• Both the request body and response body are in JSON format (JSON object).
• When the request is processed successfully, the response status code is 200, and the response body contains the following properties:

• When the request is not processed successfully, the response body contains the following properties:

Main operations provided by the service are as follows:

• infer

Performs document analysis.

POST /layout-parsing

4. Request Parameter Description

Serving:
  extra:
    max_num_input_imgs: null

Serving:
  visualize: False

• When the request is processed successfully, the result field in the response body has the following properties:

Each element in layoutParsingResults is an object with the following properties:

markdown is an object with the following properties:

5. Error Code Description

Note: If you encounter any issues during use, please feel free to submit feedback in the issue section.