.:: Previous Week ::: Next Week ::.
Week 1:
 |
| Graphics Lab: my work station is right by the cute pumpkin :) |
~ Tuesday, May 29 ~
My first work day. I was the first one in the lab at 8.30 am (an unusual event, I guess, because I was told that people are usually in the Graphics Lab early in the morning). It was a good idea to bring my laptop since I still didn't have a computer account to log into the lab PCs. I spent most of the morning working on my research website.I met with Jessica in the afternoon. We discussed who I will be working with and what my project will be about. For now, I have to do a lot of reading about Wiimotes, how they work and what people have already done with them.
I have started researching right away: I went online and found websites about Wiimotes (videos on YouTube have also been useful). Later, I got an email from Takaaki Shiratori, a post doc who is also working on this project, with links to the academic papers about interactive control of computer animations.
The day ended very nicely. Every Tuesday, Graphics people meet with each other and have a great time at this unofficial meeting. It is a good way to meet with people in the department. During each meeting somebody presents a paper they have read or their own research project. For those who are planning to present their work somewhere else these meetings are very useful in giving them a chance to practice their talk with a smaller audience and to see what kinds of questions people may ask. I enjoyed the evening, the yummy food and the new acquaintances that I made.
~ Wednesday, May 30 ~
Today I got to work with Sophie and the Motion Capture Lab's crew. The Motion Capture Lab (usually called "mocap") has 12 infrared cameras mounted around a rather large room. By placing retro-reflective markers (like the ones we have on our shoes, for example) on a person the cameras are able to detect the motion and record it via the Vicon iQ program.
I had fun putting smaller markers together to create bigger ones. Beware of Kate with double-sided scotch tape ;)) After we finished "decorating" Justin with the markers (about 40 of them), the real work began. I carefully watched how the recording of motion is done. After the data is captured, the information from all 12 cameras is reconstructed together into a 3D animation. Soon I was sitting at the computer and running the motion capturing process by myself.
~ Thursday, May 31 ~
Objective:
Use the Wii controller with at least one of the scripts written for it, such as WiiJ. WiiJ was chosen because there are detailed instructions on how to set up the controller and run the script.
Things necessary for the experiment:
* DJing software was unnecessary. Since we just needed to test the controller and see how a simple script works, the WiiJ experiment was not carried out, because of its software complexity (DJing software was more oriented to music professionals interested in creating music rather than CS geeks who just want to learn how to use a Wiimote).
Experiment:
Takaaki brought two remotes (and extra batteries!), bluetooth USB connector by Bluespoon and a driver installation CD. Even though Windows recognizes Bluespoon as a new hardware, Takaaki warned me that without additional drivers from the CD the controller may not work. I've installed both. However, when my laptop "complained" that I cannot use wireless and bluetooth at the same time, I decided to switch to one of the lab computers.
With Jessica's help we have logged in with administrative privileges to install new hardware. After logging back into my account Takaaki installed WiinRemote that allowed us to use the controller as a mouse. I found this application awkwardly designed because the cursor didn't move very fast (even after tuning up its speed in the Options menu). Also, the controller would freeze after I tried clicking on the website's button (even though left click worked on links).
I decdided to install GlovePie, a software that allows writing and running scripts for the Wiimote. After watching a video about controlling Google Earth with a custom written GlovePie script for the Wiimote, I have installed Google Earth and visited Joseph Coulston's website to download a copy of the script.
For some reason the controller wasn't initially working. After a short break, I continued exploring the Wiimotes by myself. I found out that the problem we were having occurred most likely because we used the wrong remote, one that wasn't on (remember, we had two controllers). Perhaps, in the future when we program the remotes, we need to enable the lights on the Wiimotes, so that we can tell them apart and know which remote is connected to a computer.
 |
| The Graphics Lab invaded by the high school students who came to hear Jernej's talk |
The rest of the day (from about 4 pm) I spent with high school students who visited the mocap lab and saw Jernej's presentation about his research project. Jernej (pronounced as Yerney) is from Slovenia. He is getting ready to defend his Computer Science PhD this summer. During the presentation he showed an amazing simulation on accurate sound generation.
I was also able to contribute in the discussion about graphics and animation: since the students were familiar with Alice, I showed them one of the projects that was created by students in my Narrative Programming course at the University of Hawaii, Hilo. I had fun and really enjoyed the additional information about Jernej's project, which he explained to me afterwards.
~ Friday, June 1st ~
Today was my first "official" meeting with Jessica and Takaaki. We discussed our progress at the moment. Takaaki is working on the code for the physical equations of motion. After a quick brainstorm session Jessica outlined what we know about the use of Wiimotes and came up with a couple of different areas for the project. I will be working on performance interface. The first step would be to write a program to record the acceleration data from the controller (and figure out how to capture this information for more than one Wiimote). Having this data on hand would let us use it together with mocap data to correct the noise problem.
I have a lot of researching and reading to do!
.:: Previous Week ::: Next Week ::.