DETR Table Detection
Table Transformer is a table detection model based on the DETR architecture, specifically designed to detect and recognize table structures from document images.
Downloads 17
Release Time : 9/9/2022
Model Overview
This model is primarily used to detect table regions in document images and identify table structures such as rows, columns, and cells. It combines computer vision and OCR technologies to convert tables in images into structured data.
Model Features
DETR-based Architecture
Utilizes Transformer architecture for object detection, eliminating the need for anchor box designs in traditional methods and simplifying the detection process.
Table Structure Recognition
Accurately identifies the row, column, and cell structures of tables, providing a foundation for subsequent table data extraction.
OCR Integration
Can be combined with OCR technology to convert detected tables into editable CSV formats.
Model Capabilities
Table Detection
Table Structure Recognition
Document Image Processing
Use Cases
Document Processing
PDF Table Extraction
Extracts table data from PDF documents and converts it into structured formats.
Generates editable CSV files
Scanned Document Processing
Processes scanned document images to recognize table contents.
Restores the original structure of tables
Featured Recommended AI Models
Š 2025AIbase