Day XV: High-Speed Hijinks

I took a slight gamble this morning and didn't eat breakfast, on the hope that Bob would bring us doughnuts again. My faith was rewarded, and I enjoyed a custard-filled and a jelly-filled before heading off to the lab. No one was awarded the Horse's Ass Award, which was a bit of a disappointment, but I don't think Joe was even in today, so that might be the reason.

Erika bought a couple of Where's Waldo books last night, so we had a great variety of images to work with. Thanks Erika! So our first idea was to take pictures of them with the Casio, since that worked so well with the Legos. Unfortunately, it didn't. When we attempted to put the images into our PowerPoint, they were far too dark and a bit distorted where the fold was. So, Jeff literally handed us an unopened printer/scanner and told us to use that. It worked much better, and now our PowerPoint is almost entirely complete.

Today also saw another videoconference. This time, though we were still guinea pigs, it was not an experiment. Sue, a VP student working for Pixar or something in California, needed to rehearse a presentation on her coding algorithm before giving the talk to a group out there. I hope that it does well, because the algorithm is amazing. It can code (designate fixations) videos that would take a human hours in less than a second! This would be a massive load off of anyone trying to conduct eyetracking experiments, so I hope that she's successful in refining it.

Before we went to lunch at Crossroads, Jeff called us into the conference room to discuss projects. We talked about our experiment, and he gave us two more assignments. First, he wants us to use the Casio camera to record (at ~60-300 fps) all of us interns making a variety of eye movements. (Saccades, smooth pursuit, opto-kinetic nystagmus, etc). Then, we have to put an eyetracker on someone who proceeds to experience all four observer/target combinations and demonstrates the respective eye movements. (These combinations being static/static, static/dynamic, dynamic/static, and dynamic/dynamic) We figure the best way to do this would be through watching a video clip such as a movie trailer. After a brief search period, we chose one of the previews for Kung Fu Panda. Shadoosh!!

Matt and Jane went outside after lunch to figure out the workings of the Casio, since it will be crucial to getting that task done next week. In the end, they produced a hilarious video of Jane giving a raspberry at 1200 fps. Watch it here. I hope you enjoy it as much as we did.

So, aside from all that, I managed to finish transcribing Px 21! I'm quite proud of myself. Actually, thinking about the high-speed algorithm got me thinking...what about using a transcription machine like is sometimes used for hearing-impaired people? Since checking a transcription is far faster than writing one and then checking it, we could implement one of those in the future. And since RIT has a large hearing-impaired population (does NTID ring any bells?) I don't think it would be unreasonable to assume that there might be one available on campus.

Well, if you don't know all ready, I'm going to be out for the next two weeks at Philmont Scout Ranch in New Mexico, taking a grand olde 100 mile hike through the Rockies. I'm quite looking forward to it, but at the same time I feel really bad for abandoning my labmates just as we get another two projects to do. I know they'll survive, but I wish I could help them work on it too.

Ja ne and sayonara,

AJ Lanphere