Open-source table-transformer-structure-recognition model - Extracting table structures from unstructured documents

Table Transformer Structure Recognition

Developed by microsoft

A Table Transformer model trained on the PubTables1M dataset for extracting table structures from unstructured documents

Downloads 1.2M

Release Time : 10/14/2022

Model Overview

A DETR-based table structure recognition model specifically designed to detect and extract table structures (e.g., rows, columns) from documents

Table Structure Recognition

Accurately identifies table structures in documents, including elements like rows and columns

Transformer-based

Utilizes the DETR architecture, leveraging the powerful capabilities of Transformers for object detection

Large-scale Training Data

Trained on the PubTables1M dataset, which includes a vast number of table samples

Table Detection

Table Structure Recognition

Document Analysis

Document Processing

PDF Table Extraction

Extracts table data from PDF documents

Accurately identifies table structures and content

Document Digitization

Converts tables in paper documents into structured data

Improves data entry efficiency

Data Analysis

Table Data Preprocessing

Prepares structured table data for analysis

Reduces manual processing time

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base