SUBMIT ONLINE: https://www.easychair.org/conferences/?conf=prai2026 (Please choose Special Session 17)

Document image understanding and layout analysis are fundamental technologies for digital document processing, information extraction and office intelligence. This special session centers on deep learning-based methods for document image analysis, covering layout detection, structure recognition, text extraction and content parsing for various printed, handwritten and complex documents. We invite research on innovative network architectures, multimodal fusion algorithms and practical solutions to tackle challenges like distorted documents, noisy backgrounds, diverse layouts and cross-lingual scenarios. This session brings together researchers and industry practitioners to share latest findings, technical breakthroughs and real-world applications, facilitating academic exchange and technological advancement in document image intelligence.

Chairs:
Cihui Yang Nanchang Hangkong University, China

 

RELATED TOPICS

Topics of interest include, but are not limited to:

• Deep learning for document layout detection, segmentation and reconstruction
• Text recognition, handwriting recognition and character extraction from documents
• Structure analysis of forms, invoices, receipts and historical archives
• Multimodal document understanding combining text, image and layout features
• Low-quality, distorted and degraded document image enhancement & analysis
• Cross-lingual and multilingual document image processing
• End-to-end systems for intelligent document parsing and information extraction
• Datasets, evaluation benchmarks and lightweight models for document analysis