{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/pp-structurev2-a-stronger-document-analysis","title":"PP-StructureV2: A Stronger Document Analysis System","arxiv_id":"2210.05391","date":"2022-10-11","proceeding":null,"authors":["Chenxia Li","Ruoyu Guo","Jun Zhou","Mengtao An","Yuning Du","Lingfeng Zhu","Yi Liu","Xiaoguang Hu","dianhai yu"],"abstract":"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.","url_abs":"https://arxiv.org/abs/2210.05391v2","url_pdf":"https://arxiv.org/pdf/2210.05391v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"pp-structurev2-a-stronger-document-analysis","repo_url":"https://github.com/PaddlePaddle/PaddleOCR","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"paddle","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"key-information-extraction","task_name":"Key Information Extraction"},{"task_slug":"knowledge-distillation","task_name":"Knowledge Distillation"},{"task_slug":"network-pruning","task_name":"Network Pruning"},{"task_slug":"relation-extraction","task_name":"Relation Extraction"},{"task_slug":"table-recognition","task_name":"Table Recognition"}],"methods":[{"method_slug":"knowledge-distillation","method_name":"Knowledge Distillation"},{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/network-pruning-on-cifar-100","task":"Network Pruning","dataset":"CIFAR-100","model":"+U-DML*","rank_in_archive_order":5,"of":5,"metrics":{"Inference Time (ms)":"675.56"},"uses_additional_data":false},{"leaderboard":"/sota/table-recognition-on-pubtabnet","task":"Table Recognition","dataset":"PubTabNet","model":"SLANet","rank_in_archive_order":5,"of":13,"metrics":{"TEDS (all samples)":"96.3","TEDS-Struct":"97.01"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2210.05391","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}