Scope and Motivation
Document Analysis has long suffered from both a fragmented landscape of task-specific datasets and a strong focus on narrow-focused information extraction and document conversion tasks.
The Scaling-up document Image understanding workshop aims to open the discussion on possible ways for the community to align data preparation efforts and define large-scale (grand) challenges that drive progress in the field.
Through the workshop, we aspire to contribute to the definition of grand challenges for the community. This is meant to be one of a series of such events to be organized on our scientific forums in the near future.
The Deep Document Workshop at ICFHR 2022, was our first successful event in this series. We would like to set the seed for an initiative to create our own community’s document-oriented “ImageNet”, over which multiple long-term challenges can be defined.
The workshop is meant to discuss this possibility and help us define a way forward. Come, join us to explore this together.
The workshop includes invited talks from eminent researchers and practitioners and a panel discussion.
Date: 24thAugust, 2023
Venue: Adobe World Headquarters , 345 Park Avenue , San Jose, California, USA
|Time - PDT
|09:00 M - 09:05 AM
|Welcome & Introduction - Seiichi Uchida
|09:05 AM - 09:50 AM
| Invited Talk - Yasuhisa Fujii ,
What are grand challenges in DAR? Do we already know?
|09:50 AM - 10:10 AM
|10:10 AM - 10:40 AM
|10:40 AM - 11:25 AM
| Invited Talk - J. Stephen Downie ,
Beyond OCR: Non-Textual Opportunities and Challenges at the HathiTrust Research Center
|11:25 AM - 12:25 PM
|Panel Discussion: Challenges and Datasets in DAR
|12:25 AM - 12:30 PM