Hello there, this is Fujimoto (D3).
At first, let me introduce the main motivation of this blog.
This blog aims to report lab members’ activities (e.g. conference participation)
and lab’s event.
I’m really delighted if you could know the atmosphere of our lab through this blog!
I attended the international conference about virtual reality, IEEE VR which was held in Minneapolis, Minnesota, USA. It was very cold and snowing!
I attended it both as a oral presenter and a research demonstration presenter(so hard! `_`)．
Submission : 116
Long paper :18
Short paper :10
Most conference has some sessions that a famous researcher gives a talk
, which is called “keynote”.
First keynote speaker is Prof. Henry Fuchs at North Carolina univ.
His talk is always very impressive and energetic. At this time, he mentioned about that Oculus VR, Inc. was bought by facebook inc. and this might be a big chance to appeal “Virtual Reality” to various field and markets.
Second keynote speaker, Prof. Hunter Hoffman at HIT lab told that the immersive virtual reality is valuable as an adjunctive non-pharmacologic analgesic for people with severe burn wounds suffering from uncontrolled pain during wound care.
Practical systems he mentioned made me re-recognize that VR has the various potential.
I will show you some paper I’m interested in
Mixed Reality Virtual Pets to Reduce Childhood Obesity,
Kyle Johnsen, Sun Joo Ahn, James Moore, Scott Brown, Thomas P. Robertson, Amanda Marable, and Aryabrata Basu
This research aims to reduce the high rates of childhood obesity in the developed world by virtual pet. It won the best paper award in this conference!
A mixed reality system that allowed children to exercise, play with, and train a virtual pet using their own physical activity as input was constructed.
The author conducted a user study with 60 participants and evaluated effectiveness of this system.
Global Localization from Monocular SLAM on a Mobile Phone,
Jonathan Ventura, Clemens Arth, Gerhard Reitmayr, and Dieter Schmalstieg
The author targets a hand-held device (e.g. phone with a camera) and proposes the combination of a keyframe-based monocular SLAM system and a global localization method.
The server side has large number of 3D point in advance and global localization can be conducted by 3D-3D matching between 3D point in the server and 3D point calculated by SLAM process of the client side (hand-held device).
Evaluations highlight the advantages of SLAM localization over a model-based approach which uses single-image localization on each frame.
Oral presentation for the long paper
I can not decide whether my presentation is good or not but I think it was so-so.
I always had a same feeling after the presentation but my English in the Q & A session was poor.
I wonder I may be able to answer to questions fluently if I try to think about my research topic and its logic regularly for the practice.
The proposed demonstration is the projection for a T-shirt shaped white cloth.
A retro-reflective marker pattern is attached onto the surface of the T-shirt shaped white cloth
and using this information, the texture image is projected onto the its surface so as to change the appearance of the T-shirt.
Because my system did not work properly last time, I did my best to improve the system this time.
Then system worked well and I could let many people see our demo.
Most of visitors gave us the following same comment.
“Why doesn’t this system work in real-time?”
Too many such kind of comments made me get motivated!
I will make this system work in real time for the next step, probably next time.
My impression after the conference is as follows.
As our professor always mentioned, it is crucial to continue to consider the technology is finally used for what, by whom, and how, if we develop some technologies