February 5, 2007

Notes from ISWC 2005

Notes from the 9th IEEE International Symposium on Wearable Computers. Website is here. This is really late, but it was on my to-do list and I won't cross it off without doing it.

Introduction Notes:
Opening ceremony was conducted by the Mayor of Osaka!
The conference had a 25% acceptance rate for publications
6 people nominated for best paper award.(top 10% of accepted papers)

The keynote talk was given by a guy who built a robotic suit that power assists its users. Applicable to disabled assistance, rescue work.

fascinating talk by KDDi suit
70% market saturation of cell phones
83% have 3G service.

GPSfix > 4 satellites
Hybrid -> 4 satellites plus cell tower
Indoor -> 3 or more cell towers
last -> 1 cell tower

Location based: services navigation, where am i, where is a friend, security/emergency
EZ Navi Walk -> GPS based navigation service for pedestrians
EZ "Josyuseki-Nav" Navigation for the Navigator - phone based navigation for cars plus parking availability
EZ Osagashi Navi - find lost cell phone, find friend
GPSMap -> corporate manager can monitor mulitple phones of "workers" simultaneously
Gaming
Treasure Hunt Search for treasures hidden somewhere with a phone
Tag using keitai and other people's information
Anshin Navi: Parents monitor kids location, support for boundaries and times within the boundaries.
Chaku-uta - music related services: primarily ring tones - digital rights management blah blah.
Song Search -> let phone talk listen to music and the song is searched for then you can buy it.

Combination of keitai and broadcast - see what's going on on FM radio in real-time.

handset can receive digital TV broadcasts

Contactless IC card embedded in telephone for train/bus passes or as electronic money or as an employee or student ID or as a ticket.

Interesting future of keitai - > it will bear more social responsibility for security and reliability.
He argues that American culture will not adopt keitai because we have more space!
The indoor location positioning was friggin impressive!
QVGA resolution 1/4 VGA.
PC document presentation ability -> PowerPoint, etc.

Curiously, no phones rang during the whole conference.

Dinner with Matthai, Kurt, Trevor and Adam. Raw duck, sugared french fries and ice cream.

In the morning of 10/20, a talk on trying to taxonomize information textiles. Not very compelling to me.

Then a talk on silicon skin as an intelligent textiles. Possibly cool, but not interested in this.

The Osaka ferris wheel lights up at night with a color that is based on the weather forecast for the next day.

Things about Osaka: women's train car, bike's running me over, disabled people

Next talk XS Labs (eXtraSoft Labs) worth checking out their paper on shape-changing garments. Okay totally cool - clothing that change shapes. Cool video that I need to get. Maybe on the video proceedings? From Canada

Next talk was on embroidering conductive thread through flexible pcbs for wearable computing

Next talk was on annotating objects in an Augmented Reality system by using depth cues so that a person could stand in one location and annotate a number of different objects in the world without having to go to them.

The next talk was on an event-driven system for navigation with wearable computing. The API/event model proposed had no accommodation for uncertainty though.

Next talk on artificial synesthesia for wearable computing. The video looked cool because he recolored digits on a sheet of paper with different colors, but there are questions about scalability to general tasks.

Lunch provided by conference.

After lunch there was a guy trying to do "Experience Capture" using a necklace with a suite of sensors on it, like a camera.

Then wearable hand activity recognition for Event Summarization: Used computer vision to learn activities from hand motions, kind of crazy... The most interesting aspect of this work was that they used an active camera that was worn.

Then a paper on detecting conversations with wearables.

A paper on using wearable technology for crime scene investigation.

There is a theme hear of annotating things in space using wearable computers.

Another active camera demo during the demo session. It was coupled with a laser pointer to enable remote collaboration. Kind of cool.

Tom Martin at Virginia Tech was looking at denial of service attacks that focussed on attacks which drain the power of mobile devices. in that conversation he told me that there are security attacks in which a user who has access to power consumption data and has a sequence of encryption events with them can deduce the majority of bits of a DES key. Very cool.

10/21
new day

Wearable Full-body Motion Capture System: self-contained, textile shape measurement. Study was simulation based on motion capture data for placement of sensors. Hard to get really excited about this because it was based on motion capture dataT

Next paper using ultrasonic hand tracking to augment activity recognition. Bat->cricket->University of Bristol -> Hexamite. Hand tracking requires finer resolution. Pre-segmented data for recognition. No good ideas for the "other" category.

Detection of eating and drinking gestures using inertial body-worn sensors. This is supposed to support figuring out what you are eating, but I don't quite get this.

Good paper on detection stimming behavior in autistic kids.

Two papers on input device technologies like Twiddlers.

Stuff that sucked was projects ideas with no user evaluation -> all of them had been done before in some manner or another and added little value to the research community. User data was fascinating and insightful though.

Posted by djp3 at 11:45 AM | Comments (0) | TrackBack (0)