PP-StructureV2: A Stronger Document Analysis System

by   Chenxia Li, et al.

A large amount of document data exists in unstructured form such as raw images without any text information. Designing a practical document image analysis system is a meaningful but challenging task. In previous work, we proposed an intelligent document analysis system PP-Structure. In order to further upgrade the function and performance of PP-Structure, we propose PP-StructureV2 in this work, which contains two subsystems: Layout Information Extraction and Key Information Extraction. Firstly, we integrate Image Direction Correction module and Layout Restoration module to enhance the functionality of the system. Secondly, 8 practical strategies are utilized in PP-StructureV2 for better performance. For Layout Analysis model, we introduce ultra light-weight detector PP-PicoDet and knowledge distillation algorithm FGD for model lightweighting, which increased the inference speed by 11 times with comparable mAP. For Table Recognition model, we utilize PP-LCNet, CSP-PAN and SLAHead to optimize the backbone module, feature fusion module and decoding module, respectively, which improved the table structure accuracy by 6% with comparable inference speed. For Key Information Extraction model, we introduce VI-LayoutXLM which is a visual-feature independent LayoutXLM architecture, TB-YX sorting algorithm and U-DML knowledge distillation algorithm, which brought 2.8% and 9.1% improvement respectively on the Hmean of Semantic Entity Recognition and Relation Extraction tasks. All the above mentioned models and code are open-sourced in the GitHub repository PaddleOCR.


page 3

page 4


Document-Level Relation Extraction with Adaptive Focal Loss and Knowledge Distillation

Document-level Relation Extraction (DocRE) is a more challenging task co...

PP-OCRv3: More Attempts for the Improvement of Ultra Lightweight OCR System

Optical character recognition (OCR) technology has been widely used in v...

A Trigger-Sense Memory Flow Framework for Joint Entity and Relation Extraction

Joint entity and relation extraction framework constructs a unified mode...

Document Layout Analysis via Dynamic Residual Feature Fusion

The document layout analysis (DLA) aims to split the document image into...

PP-OCRv2: Bag of Tricks for Ultra Lightweight OCR System

Optical Character Recognition (OCR) systems have been widely used in var...

Unimodal and Multimodal Representation Training for Relation Extraction

Multimodal integration of text, layout and visual information has achiev...

Please sign up or login with your details

Forgot password? Click here to reset