File size: 500 Bytes
c9f61a8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
---
title: LayoutLMv3 Document Extraction
emoji: 📄
colorFrom: blue
colorTo: green
sdk: docker
app_file: app.py
pinned: false
---

# LayoutLMv3 Document Extraction with OCR

This Space provides document extraction with bounding boxes using LayoutLMv3 and Tesseract OCR.

## Features
- PDF processing with OCR
- Image text extraction
- Bounding box coordinates for each text element
- Multi-page PDF support

## API Usage

Send POST requests to `/extract` with base64-encoded PDF or image:
```python