You will find below the preliminary program.
A pdf version of the program is available.
Speakers are invited to read the guidelines for speakers in order to prepare their presentation.
Date : Thursday 24th August
Location
GREC 2023 will be held at the Adobe World Headquarters
345 Park Avenue, San Jose, California, USA
Morning
| 9:00-9:10 | Opening session General Chair : Jean-Christophe Burie |
| 9:10-10:30 | Keynote Talk Chair : Jean-Christophe Burie Professor Richard Zanibbi Rochester Institute of Technology Title : Math Formula Recognition and Search in MathDeck Abstract : MathDeck is a new prototype for searching PDF documents using both math formulas and text (https://demo.mathdeck.org). The first version of the prototype indexes research papers on Natural Language Processing in the ACL Anthology (https://aclanthology.org). In this talk, we will discuss the design of MathDeck, focusing on modules that recognize, index, and search math formulas whether they are in ‘born digital’ vector graphics representations (i.e., character and line drawing instructions) or in raster images. We will also talk more broadly about the role of graphics recognition in Mathematical Information Retrieval, and research opportunities including multi-modal search engines that link documents, slides, audio, and videos through their graphics and text. This work has been supported by the National Science Foundation (USA) and the Alfred P. Sloan Foundation. |
| 10:30-11:00 | Coffee-break |
| 11:00-12:30 | Session 1 : Symbol recognition Chair : Bart Lamiroy |
| (11:00-11:15) Segmentation-Free Alignment of Arbitrary Symbol Transcripts to Images Pau Torras, Mohamed Ali Souibgui, Jialuo Chen, Sanket Biswas, Alicia Fornes (11:15-11:30) Can Pre-trained Language Models help in Understanding Handwritten Symbols? Adarsh Tiwari, Sanket Biswas, Josep Llados (11:30-11:45) Text Extraction for Handwritten Circuit Diagram Images Johannes Bayer, Shabi Haider Turabi, Andreas Dengel (11:45-12:00) Optical Music Recognition: Recent Advances, Current Challenges, and Future Directions Jorge Calvo-Zaragoza, Juan Carlos Martinez-Sevilla, Carlos Penarrubia, Antonio Ríos-Vila (12:00-12:30) Panel Discussion | |
| 12:30-13:30 | Lunch |
Afternoon
| 13:30-14:40 | Session 2 : Comics Analysis Chair : Rafael Lins |
| (13:30-13:45) KangaiSet: A dataset for visual emotion recognition on mangas Ruddy Théodose, Jean-Christophe Burie (13:45-14:00) Automatic Detection of Comic Characters: An Analysis of Model Robustness Across Domains Javier Lucas Aura, Antonio Javier Gallego, Jorge Calvo-Zaragoza (14:00-14:15) Detection of buried complex text. Case of Onomatopoeia in comics books John Benson LOUIS, Jean-Christophe Burie (14:15-14:40) Panel Discussion | |
| 14:40 – 15:10 | General Discussion : What is the future of pure research in graphics recognition ? Chair : Jean-Christophe Burie |
| 15:10-15:40 | Coffee Break |
| 15:40-17:10 | Session 3 : Graphics Chair : Josep Llados |
| (15:40-15:55) Reconstruction of Power Lines from Point Clouds Alexander Gribov, Khalid Duri (15:55-16:10) A Survey and Approach to Chart Classification Anurag Dhote, Mohammed Javed, David Doermann (16:10-16:25) FPNet: Deep Attention Network for Automated Floor Plan Analysis Abhinav Upadhyay, Alpana Dubey, Suma Mani Kuriakose (16:25-16:40) MuraNet: Multi-task Floor Plan Recognition With Relation Attention Lingxiao Huang, Jung-Hsuan Wu, Chiching Wei, Wilson Li (16:40-17:10) Panel Discussion | |
| 17:10-17:25 | Conclusion and Closing General Chair : Jean-Christophe Burie |