GLESDO2021: Graph representation Learning for Scanned Document analysis Lausanne, Switzerland, September 5-10, 2021 |
Conference website | https://www.glesdo-icdar2021.ml/ |
Submission link | https://easychair.org/conferences/?conf=glesdo2021 |
Context
Robust reading, also known as automatic document image processing, is an essential task in various applications areas such as data invoice extraction, subject review, medical prescription analysis, etc. and holds significant commercial potential. Several approaches are proposed in the literature, but datasets' availability and data privacy challenge it.
Considering the problem of information extraction from documents, different aspects must be taken into account, such as (1) document classification (2) text localization (3) OCR (Optical Character Recognition) (4) table extraction (5) key information detection. In this context, the graph-based approaches are attractive methods for document processing. In fact, graphs are a natural way to represent the connections among objects (text, blocks, images, etc.) and aim to discover novel and hidden knowledge from data. The extracted text from scanned documents can be represented in the shape of a graph to exploit the best features of their characteristics. On the other hand, understanding spatial relationships is critical for text document extraction results for some applications such as invoice analysis. The aim is to capture the structural connections between keywords (invoice number, date, amounts) and the main value (the desired information). An effective approach requires a combination of spatial and textual information.
Objective
This workshop aims to bring together an area for experts from industry, science, and academia to exchange ideas and discuss on-going research in graph representation learning for scanned document analysis.
We encourage the description of novel problems or applications for document image analysis in the area of information retrieval that have emerged in recent years. Furthermore, we also encourage works that develop new scanned document datasets for novel applications.
Topics of interests:
We invite the submission of original works that is related -- but are not limited to -- the topics below:
● Graph embeddings
● Deep learning for graph
● Probabilistic graphical models for graphs
● Graph-based approaches for text mining
● Graph-based approaches for spatial component in scanned document
● Graph representation learning for NLP
● Graph-based approaches using kernels
● Spectral graph clustering
● Semi-supervised graph-based methods
● Dynamic graph analysis
● Information Retrieval and Extraction using Graph-based methods
● Knowledge graph for semantic document analysis
● Semantic understanding of document content
● Entity and link prediction in graphs
● Merging ontologies with graph-based methods using NLP techniques
● Cleansing and image enhancement techniques for scanned document
● Document structure and layout learning
● OCR based graph methods
● Font text recognition in scanned document
● Table identication and extraction from scanned documents
● Handwriting detection and recognition in documents
● Signature detection and verication in documents
● Visual document structure understanding
● Visual Question Answering
● Invoice analysis
● Scanned documents classification
● Scanned documents summarization
● Scanned documents translation
Organizing committee
Rim Hantach, Engie, France
Rafika Boutalbi, Trinov, France - Stuttgart University, Germany
Philippe Calvez, Engie, France
Balsam Ajib, Trinov, France
Thibault Defourneau, Trinov, France