P

PP DocLayout L

Developed by PaddlePaddle
PP-DocLayout-L is a high-precision document layout region localization model based on the RT-DETR-L architecture, supporting the detection of 23 common document layout categories.
Downloads 285
Release Time : 6/6/2025

Model Overview

This model is specifically designed for document layout detection and can accurately locate various elements in the document, such as text, tables, images, etc., providing structured information for document processing.

Model Features

High-precision layout detection
Trained on a self-built dataset, the mAP(0.5) reaches 90.4%, enabling accurate identification of various layout elements in the document.
Multi-category support
Supports the detection of 23 common document layout categories, including text, tables, images, headers and footers, etc.
Applicable to a wide range of document types
Applicable to various document types such as papers, magazines, contracts, books, test papers, and research reports.

Model Capabilities

Document layout detection
Table recognition
Text region localization
Image region detection

Use Cases

Document processing
Academic paper analysis
Automatically identify structural elements such as titles, main texts, and references in papers
Realize automatic classification and structuring of paper content
Contract parsing
Detect key clauses and signature areas in contract documents
Improve the efficiency of contract review
Office automation
Document digitization
Convert scanned documents into structured digital formats
Facilitate subsequent content retrieval and processing
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase