Workshop on Industrial Applications of Document Analysis and Recognition

In conjunction with ICDAR 2019, Sydney Australia
23-25 September 2019


Call for Papers



ICDAR Main Web


This workshop invites researchers and engineers from industry, or public administrations, or universities in partnership with industry or administrations, to briefly present their applied research results and innovative applications.

There is a lot of innovative work being conducted in industry, or in public administrations, or in universities in partnership with industry or administrations. Usually, the aim of applied research for industrial innovation is to propose solutions to breakthrough two main dimensions: the application scope (application to a new domain, new uses, ability to adapt the application to another domain with a null or minimized effort) and the performance (highest success with lowest critical error, resistance on specific constraints).

Most applied research involves reuse, combine, tune and train existing methods with the advantage of accessing large corpus of data. Industrial researchers either adapt or create methods with scientific innovations, but also with heuristics and, whatever can pragmatically reach the target for an acceptable scope. Usually, that applied research would not be accepted at ICDAR because the lack of methodology novelty. Furthermore, when industrial researchers create a real novel approach, the business strategy in the industrial competition prevents the disclosure of the information. Hence, that applied research often is not submitted to ICDAR.

However, all these works are still of interest to attendees of ICDAR, because experiments with large real-life dataset provides pragmatic feedbacks on the technologies. It really demonstrates the state-of-the-art limitations, it infers a ranking of technology performance and, both workarounds or optimizations may point out some important research perspectives. In addition, it highlights end-user problems and needs, i.e. it may inform academic perspectives. We do believe that, even if an application is not scientifically original or not clearly described, the result status and discussions around this method have a strong value for ICDAR attendees.

This workshop is an effort to make information about innovative applications available to the general ICDAR attendee, and to encourage and provide those researchers and engineers with the opportunity to attend ICDAR. We also hope this will build more interaction between industry/public administration and academia.

Topics of Interest:

Topics of Interest include, but are not limited to:
  • Applications of document analysis
  • Document analysis systems
  • Document Image Processing
  • Physical and logical layout analysis
  • Character and text recognition
  • Pen-based document analysis
  • Historical document analysis
  • Symbol and graphics recognition
  • Document forensics
  • Human document interaction
  • Scene text detection and recognition
  • Document retrieval
  • verification and writer identification
  • Multimedia documents
  • Performance evaluation
  • Machine learning for document analysis
  • Cognitive issues of documents
  • Semantic information extraction from documents
  • Document summarization classification and translation
  • Document simulation and synthesis
Acceptance will be based on the criteria:
  • originality of the problem
  • pertinence of the evaluation protocol
  • pertinence of the performance compared to business competition
  • pertinence of the discussion
  • impact : genericity and adaptability of the approach, i.e. adaption effort required when facing document variability or adaptation effort to setup a new user case.
The weighting of the originality of the method will decrease if the work is applied to a new scenario with high performance, and if the method demonstrates good adaptability.

Workshop on Industrial Applications of Document Analysis and Recognition - WIADAR 2019