Experience in ACM Intelligent Virtual Agent (IVA2019) Paris, France

My name is Rex Hsieh, an intern at GREE VR Studio Lab and a 3rd year doctoral student in Kanagawa Institute of Technology (KAIT).

A photo in VRSionUp#4, workshop of GREE VR Studio Lab, I’m locating on center right.

On June 29th 2019, I embarked on a journey to Paris, France to present my poster submission titled: “Evaluation of Avatar and Voice Transform in Programming E-Learning Lectures” at a long established international conference “Intelligent Virtual Agent(IVA2019)”. IVA was hosted at Centre national de la recherche scientifique (CNRS) this year jointly with Computer Animation and Social Agents (CASA2019) at the same venue. Together these 2 conferences about avatar and social agent research lasted from July 1st to July 5th with CASA from July 1st to 4th and IVA from July 2nd to 5th.

Registration for CASA began at 8:30 AM in CNRS where conference attendees first picked up badges from booth manned by local volunteers, many of whom students and researchers themselves. The conference began from 9:30 AM to 17:00 PM everyday composed of 8 sessions with 2 sessions taking place at the same time and attendees had to choose between 1 of the 2 sessions. I’ve joined to “Session 4: Virtual Reality I” chaired by Prof. Masaki Oshita. My favorite presentation in this session was “Human-Virtual Character Interaction: Towards Understanding the Influence of Haptic Feedback” by Claudia Krogmeier, Christos Mousas, and David Whittinghill. This research evaluates the influence of haptic feedback in VR environment by having virtual avatars bump into users and generating a wide range of haptic feedbacks. The paper not only uses Galvantic Skin Response (GRS) to gather physiological data but also measures elements such as presence, embodiment, positive effect, negative effect, and realism from data obtained using Likert Scale survey questions. The result is significant differences were found in embodiment, realism of virtual character interaction, and haptic feedback realism. Due to how solid the research method and conclusion are I picked this paper as my favorite presentation out of “Session 4: Virtual Reality I”.

From “Human–virtual character interaction: Toward understanding the influence of haptic feedback” (Link)

Another equally well conducted experiment from this session goes to “Virtual titration laboratory experiment with differentiated instruction” by Bo-Jyun Wu, Sai-Keung Wong, and Ta-Wei Li. This research tests the effectiveness of a virtual reality chemistry lab on both people with chemistry and no chemistry backgrounds and utilized leap motion to detect users’ hand gestures during titration experiments. Assistance tools for learning and operating virtual items are also provided. In conclusion, the research team found that VR chemistry environment as able to boost the confidence of people with no chemistry background but due to a few technical shortcomings such as leap motion unable to detect some finger movements this titration VR education environment is not perfect. During the entire presentation I was inspired by the clarity and honesty of the presenter and plans on using the knowledge learned to refine my research.

Catch up from Keynotes

The keynote speakers of both conferences were invited from organizations such as Disney Research Studios and UNESCO which was situated right at the heart of Paris close to the Eiffel tower. The keynote speech given by Thabo Beeler from Disney Research Studios concerns the different facial capture approaches he developed which acquires data from specific facial features and then have the captured data transferred to the other departments where further digital touch ups and manipulations are done. In a paper he mentioned during the keynote speech titled: “Appearance Capture and Modeling of Human Teeth”, Thabo Beeler described in detail how he was able to capture the appearance of teeth and visualize them digitally on movie sets. Below is a video from another one of Thao Beeler titled: “Practical Dynamic Facial Appearance Modeling and Acquisition”.

Official Link of Disney Research Studios
Keynote Speech on day 1 of CASA by Thabo Beeler who works at the Disney Research Studios.

From UNESCO we have Mark West who gave a speech concerning the issue of gender gap in tech industry. Specifically Mark mentioned an articles published by UNESCO titled: “I’d blush if I could: closing gender divides in digital skills through education which dwelves deep into the issue of inequality in technology education in both the developed and developing countries. Other points brought out by Mark West include how AI programs containing an audio compartment such as Siri, Cortana, and Alexa were built with a female’s audio and personality in mind which leads to “abuses” of female sounding programs that can eventually lead to desensitization of abuses towards females.

Lunch session in French conference

There was a lunch session at 13:00 PM where attendees are given 1 lunch and cafe ticket per day.

Lunch at CNRS composed of 2 deserts, 1 drink, and 1 main dish. It may be a common “cantine” in France.

Aside from the sessions, the conference is also composed of keynote sessions where accomplished researchers from both the academia and industry were invited to give talks concerning their research and their view points on issues faced by the industry.

“Evaluation of Avatar and Voice Transform in Programming E-Learning Lectures” Poster Session

My poster pitch was given on July 4th on Thursday. At 12:00 pm, the 19 poster and demo presenters lined up at the auditorium and each was given 1 minute to pitch their research. I was given number 7 and while I was a bit nervous on the stage, the 1 minute speech was successfully delivered and I was even able to link my research back to a point made by Mark West, and that being instead of automatically assume all users want a female avatar, it is the best to give them choices from a range of avatars of different genders. By linking my topic with a point made by a keynote speaker, I was able to demonstrate the strengths and charactersitics of my reserach and therefore grab the audiences’ interest. Indeed many audiences returned later at the poster demonstration session to talk with me about my research.

Rex Hsieh’s 1 Minute Poster Pitch Material.

IVA/CASA Conference Conclusion

As a whole my presentation in CASA and IVA was pretty well received and I was able to obtain valuable information on how to advance further with my research. Furthermore by absorbing research methods and techniques mentioned by world class researchers from all across the world, I have gained an understanding on how to advance further with my research.

I will keep on focusing on my research and strive to take this subject to the next level. Please stay tunned to my future progress as I will be attending SIGGRAPH Los Angeles 2019 soon later this July and will be sure to give another update.

82. Effectiveness of Facial Animated Avatar and Transformed Voice in eLearning Programming Course

Representative image from “Effectiveness of Facial Animated Avatar and Transformed Voice in eLearning Programming Course



See you there!

Rex HSIEH — 27th July, 2019

R&D division for Creating a new future with XR entertainment research and VTuber industry with REALITY, Inc. in GREE Family.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store