Document Analysis and Recognition - ICDAR 2021: 16th International Conference, Lausanne, Switzerland, September 5-10, 2021, Proceedings, Part III

by Josep Lladós (Editor), Daniel Lopresti (Editor), Seiichi Uchida (Editor)

Overview

This four-volume set of LNCS 12821, LNCS 12822, LNCS 12823 and LNCS 12824, constitutes the refereed proceedings of the 16^th International Conference on Document Analysis and Recognition, ICDAR 2021, held in Lausanne, Switzerland in September 2021. The 182 full papers were carefully reviewed and selected from 340 submissions, and are presented with 13 competition reports.

The papers are organized into the following topical sections: extracting document semantics, text and symbol recognition, document analysis systems, office automation, signature verification, document forensics and provenance analysis, pen-based document analysis, human document interaction, document synthesis, and graphs recognition.

Product Details

ISBN-13:	9783030863333
Publisher:	Springer International Publishing
Publication date:	09/06/2021
Series:	Lecture Notes in Computer Science , #12823
Edition description:	1st ed. 2021
Pages:	756
Product dimensions:	6.10(w) x 9.25(h) x (d)

Table of Contents

Extracting Document Semantics.- MiikeMineStamps: A Long-Tailed Dataset of Japanese Stamps via Active Learning.- Deep Learning for Document Layout Generation: A First Reproducible Quantitative Evaluation and a Baseline Model.- Text and Symbol Recognition.- MRD: A Memory Relation Decoder for Online Handwritten Mathematical Expression Recognition.-Full Page Handwriting Recognition via Image to Sequence Extraction.- SPAN: a Simple Predict & Align Network for Handwritten Paragraph Recognition.- IHR-NomDB: The Old Degraded Vietnamese Handwritten Script Archive Database.- Sequence Learning Model for Syllables Recognition Arranged in Two Dimensions.- Transformer for Handwritten Text Recognition using Bidirectional Post-Decoding.- Zero-Shot Chinese Text Recognition via Matching Class Embedding.- Text-conditioned Character Segmentation for CTC-based Text Recognition.-Towards Fast, Accurate and Compact Online Handwritten Chinese Text Recognition.- HCADecoder: A Hybrid CTC-Attention Decoder for Chinese Text Recognition.-Meta-learning of Pooling Layers for Character Recognition.- Document Analysis Systems.- Text-line-up: Don’t Worry about the Caret.- Multimodal Attention-based Learning for Imbalanced Corporate Documents Classification.- Light-weight Document Image Cleanup using Perceptual Loss.- Office Automation.- A New Semi-Automatic Annotation Model via Semantic Boundary Estimation for Scene Text Detection.- Searching from the Prediction of Visual and Language Model for Handwritten Chinese Text Recognition.- Towards an IMU-based Pen Online Handwriting Recognizer.- Signature Verification.- 2D vs 3D online writer identification: a comparative study.- A Handwritten Signature Segmentation Approach for Multi-resolution and Complex Documents Acquired by Multiple Sources.- Attention based Multiple Siamese Network for Offline Signature Verification.- Attention to Warp: Deep Metric Learning for Multivariate Time Series.- Document Forensics and Provenance Analysis.- Customizable Camera Verification for Media Forensic.- Density Parameters of Handwriting in Schizophrenia and Affective Disorders Assessed Using the Raygraf Computer Software.- Pen-based Document Analysis.- Language-Independent Bimodal System for Early Parkinson’s Disease Detection.-TRACE: A Differentiable Approach to Line-level Stroke Recovery for Offline Handwritten Text.- Segmentation and graph matching for online analysis of student arithmetic operations.- Applying End-to-end Trainable Approach on Stroke Extraction in Handwritten Math Expressions Images.- A Novel Sigma-Lognormal Parameter Extractor for Online Signatures.- Human Document Interaction.- Near-perfect Relation Extraction from Family Books.- Estimating Human Legibility in Historic Manuscript Images - A Baseline.- A Modular and Automated Annotation Platform for Handwritings: Evaluation on Under-resourced Languages.- Reducing the Human Effort in Text Line Segmentation for Historical Documents.- DSCNN: Dimension Separable Convolutional Neural Networks for character recognition based on inertial sensor signal.- Document Synthesis.- DocSynth: A Layout Guided Approach for Controllable Document Image Synthesis.- Font Style that Fits an Image — Font Generation Based on Image Context.- Bayesian Hyperparameter optimization of Deep Neural Network algorithms based on Ant Colony optimization.- End-to-End Approach for Recognition of Historical Digit Strings.- Generating Synthetic Handwritten Historical Documents With OCR Constrained GANs.- Synthesizing Training Data for Handwritten Music Recognition.- Towards Book Cover Design via Layout Graphs.- Graphics Recognition.- Complete Optical Music Recognition via Agnostic Transcription and Machine Translation.- Improving Machine Understanding of Human Intent in Charts.- DeMatch: Towards Understanding the Panel of Chart Documents.- Sequential Next-Symbol Prediction for Optical Music Recognition.- Which Parts Determine the Impression of the Font?.- Impressions2Font: Generating Fonts by Specifying Impressions.