IINAE 2024: The 6^th Workshop on Innovation Initiatives in NLP/AI/Education

Valerie Sutton

Center for Sutton Movement Writing, Inc, United States

Sutton SignWriting 2020-2050

Bringing Together Communities of Southeast Asian Sign Language Users Through Writing Their Sign Languages

• Abstract:

Sutton SignWriting is 50 years old in 2024. It is an internationally-used writing system for writing the movements of all sign languages. The writing system can be applied to writing any sign language in the world because it writes body movement. There are 40 to 60 countries writing their sign languages using SignWriting. The only requirement is that the writers know their own sign language, and if they know how to sign in their sign language, then they can learn to use the SignWriting symbols to write their sign language. SignWriting can be written by hand, or written by computer. The computer software is constantly under development. The biggest challenges in 2025-2050 is for the Sutton SignWriting system to establish a world-wide server that can handle all the computer entries made simultaneously by writers around the world. In 2024, the best software for creating dictionaries online is called “SignPuddle 2”, developed by Steve Slevinski. In this talk, I will teach you how to use SignPuddle 2 for creating a Myanmar Sign Language Dictionary.

• Biography:

Born in New York City in 1951, Valerie Sutton is an American who invented Sutton Movement Writing, a writing system for recording all body movement. Sutton Movement Writing has five sections, including: SignWriting and DanceWriting. For more detailed information about Valerie Sutton, go to valeriesutton.org or on Wikipedia:

Prarinya Siritanawan, PhD

Associate Professor at Faculty of Engineering, Shinshu University, Japan

Security in Image Recognition Systems: Understanding Adversarial Examples and their Countermeasures

• Abstract:

This talk addresses the security challenges posed by adversarial examples in image recognition systems, in particular adversarial patches that can attack image recognition in physical space. We have explored the image features behind these attacks and found the unique characteristics of adversarial patches, in contrast to common misconceptions in the field. The talk will also cover key defence strategies to improve the robustness of image recognition systems. By analysing both vulnerabilities and countermeasures, this talk will feature the importance of secure and trustworthy image recognition in the next generation of AI applications.

• Biography:

Prarinya Siritanawan is a faculty member at the Graduate School of Science and Technology, Shinshu University, Japan, where he leads a research group focused on Trustworthy Human-Centric Image Information Engineering. He received his B.Eng. and M.Eng. degrees from the Sirindhorn International Institute of Technology, Thammasat University, and his Ph.D. in Information Science from the Japan Advanced Institute of Science and Technology (JAIST). Previously, he was a Research Fellow at Nanyang Technological University, Singapore, and later an Assistant Professor at JAIST. He is a member of IEEE Society and served as the committee and a reviewer on several international conferences and journals in the areas of in computer vision, machine learning, and robotics.