Open-source table detection model table-transformer-detection - Accurately extract tables from unstructured documents

Table Transformer Detection

Developed by microsoft

A table detection model based on the DETR architecture, specifically designed for extracting tables from unstructured documents

Object Detection

Transformers

Open Source License:MIT #Document Table Detection #DETR Architecture #Unstructured Document Processing

Downloads 2.6M

Release Time : 10/14/2022

Model Overview

This model is trained on the PubTables1M dataset for detecting table structures in documents, suitable for document processing and table extraction tasks

Model Features

Based on DETR Architecture

An object detection model using Transformer architecture, with efficient table detection capabilities

Optimized for Table Detection

Fine-tuned on the PubTables1M dataset, specifically designed for table detection tasks in documents

Pre-Normalization Setting

Applies layer normalization before self-attention and cross-attention to improve model stability

Model Capabilities

Document Table Detection

Table Structure Recognition

Unstructured Document Processing

Use Cases

Document Processing

Invoice Processing

Automatically detect and extract table data from invoice documents

Accurately identifies table regions in invoices

Report Analysis

Extract table data from business reports for analysis

Capable of handling complex report document formats

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご

Table Transformer Detection

Model Overview

Model Features

Model Capabilities

Use Cases

🚀 Table Transformer (fine-tuned for Table Detection)

🚀 Quick Start

✨ Features

💻 Usage Examples

Basic Usage

📄 License