Description
Book SynopsisThis two-volume set LNCS 14193-14194 constitutes the proceedings of International Workshops co-located with the 17th International Conference on Document Analysis and Recognition, ICDAR 2023, held in San José, CA, USA, during August 21–26, 2023.
The total of 43 regular papers presented in this book were carefully selected from 60 submissions.
Part I contains 22 regular papers that stem from the following workshops:
ICDAR 2023 Workshop on Computational Paleography (IWCP);
ICDAR 2023 Workshop on Camera-Based Document Analysis and Recognition (CBDAR);
ICDAR 2023 International Workshop on Graphics Recognition (GREC);
ICDAR 2023 Workshop on Automatically Domain-Adapted and Personalized Document Analysis (ADAPDA);
Part II contains 21 regular papers that stem from the following workshops:
ICDAR 2023 Workshop on Machine Vision and NLP for Document Analysis (VINALDO);
ICDAR 2023 International Workshop on Machine Learning (WML).
Table of ContentsTypefaces and Ligatures in Printed Arabic Text: A Deep Learning-Based OCR Perspective.- Leveraging Knowledge Graph Embeddings to Enhance Contextual Representations for Relation Extraction.- Extracting Key-Value Pairs in Business Documents.- Long-Range Transformer Architectures for Document Understanding.-Pre-training transformers for Corporate Documents Understanding.- Transformer-Based Neural Machine Translation for Post-OCR Error Correction in Cursive Text.- Arxiv Tables: Document Understanding Challenge Linking Texts and Tables.- Subgraph-Induced Extraction Technique for Information (SETI) from Administrative Documents.- Document Layout Annotation: Database and Benchmark in the Domain of Public Affairs.- A Clustering Approach Combining Lines and Text Detection for Table Extraction.- Absformer: Transformer-Based Model for Unsupervised Multi-Document Abstractive Summarization.- A Comparison of Demographic Attributes Detection from Handwriting Based on Traditional and Deep Learning Methods.- A New Optimization Approach to Improve an Ensemble Learning Model: Application to Persian/Arabic Handwritten Character Recognition.- BN-DRISHTI: Bangla Document Recognition Through Instance-level Segmentation of Handwritten Text Images.- Text Line Detection and Recognition of Greek Polytonic Documents.- A Comprehensive Handwritten Paragraph Text Recognition System: LexiconNet.- Local Style Awareness of Font Images.- Fourier Feature-Based CBAM and Vision Transformer for Text Detection in Drone Images.- Document Binarization with Quaternionic Double Discriminator Generative Adversarial Network.- Crosslingual Handwritten Text Generation Using GANs.- Knowledge Integration inside Multitask Network for Analysis of Unseen ID Types.
