Hi guys, it's been a long time since you last heard from me. Patrick's been covering for us while we were at SCTI but now that we're separated, we all have to record our individual adventures and travels. I moved to Tokyo on Saturday 7/7/07 to a one-room mansion in Hongo. I will post some pictures of it later, but it is what you'd expect from a room in Tokyo, small and spare but functional. The first night, Jason stayed over because he went to a concert in Tokyo (with Linkin Park and Rihanna among others) . I fit him in the one-person sized space on the floor next to my bed. The following night, Michael stayed over because he had just come back from his brother's wedding in Irvine and I did the same thing for him. It worked out well because I could provide him accommodations and he could tie my tie for me the next day :).
My internship is at a robotics group in Tokyo University. My professor is Arai Tamio (last name Arai) and his group is split up between him and 2 associate professors, Yokoi and Ota. I am in Arai's group working in the Robocup project, but unfortunately not directly on the AIBO robots who play soccer for Robocup. I'm currently assigned to the overhead fisheye camera that monitors the field. It's a new camera and the only existing code that has been written simply grabs a frame and stores it. He wants me to use the camera to localize objects on the field in 2D, which sounds relatively straightforward. However, calibrating the distortion caused by the fisheye camera remains a challenge. I'm currently on my 4th day here, where the first day consisted mainly of introductions and a meeting, the 2nd day was software installation and getting familiar with the subject, and the 3rd day was a group meeting and leaving early to meet up with friends. So, I haven't done any real work yet. Contrary to what I believed coming in here, there really aren't high expectations for me despite the fact that I'm the first undergraduate intern and the faculty literally had to change the system to let me in (Arai sensei keeps telling me how surprised he is that they would allow it). He just wants me to be a "powerful support" to my colleagues, so any real work that I do will just be a nice bonus. As for the other people here... they do seem pretty hard working in general. I was really impressed on the first day because they stayed until 12 midnight, but as I realized the following day, it is because they don't get in the office until 12 noon. So, depending on whether I want my job to sound easy or hard, I can say I don't leave until midnight, or alternatively that I don't get in until noon.
Yesterday I met up with my quintet from freshman year. As luck would have it, 4 of the 5 of us are in Tokyo at the moment. Besides me, Guson (piano) is on vacation before he starts med school, Naoya (cello, just admitted to Juilliard) is doing a consulting internship at Bain, and Iris (violin, UCSD med school) is playing in a music festival in Japan. I first met up with Naoya and Guson in Shinjuku and we did some window shopping and chatted before eating a really lame dinner. It sounded good, beef tongue set, but the set was really lame, just 3 pieces of beef tongue, a bowl of rice, some pickled cabbage that tasted disgusting, and a bowl of soup. The tongue was really tough - in fact the only redeeming quality was the soup was good. And it wasn't cheap either, it was Y1450. Afterwards we met up with Iris at about 10:00 because her concert ended at 9:30. We talked for a bit in a cafe and then ate some Hokkaido ramen. Iris kept reverting back to Japanese because she didn't speak anything but Japanese until she was 5 years old. Naoya is also fluent at Japanese despite being Japanese-American. We had a nice time catching up but in the end, we stayed too late. We didn't leave the ramen place until 12, and though everybody else got back OK, I was stranded on a train 2 stops from Shibuya and had to take a taxi back home. It ended up being Y3620 to get back, which I justified as 1) you have to do it at least once to know how much it hurts and 2) it's cheaper than a capsule hotel and I wouldn't have to worry about getting back to Tokyo University the next day.
Subscribe to:
Post Comments (Atom)
3 comments:
you must've had a rough night but it sounds like it was fun. wow.. med school, internship at Bain, and Juilliard admittance.. you hang with a rather elite crowd. heh. nice to hear an update from you. can't wait to see the pictures.
I would add one more reason for you to justify the taxi expanse - the friendship and the wonderful moment together with your wonderful friends worth more than Y3620.
This is really a interesting coincident that you are assigned to work on processing a 2D image from a dome shaped lens. I have been thinking of that technique for more than 3 years, which is a part of network security monitor system. I want to replace the swing motion camera (to cut cost) with a camera use dome shaped lens. Using software to restore the distorted image back to flate 2D wide area image. This technique has a lot of commercial application. eg. surgeon's tube shaped optical inspection probe.
In your application, you may have one more level of difficulty that is you need to keep the accuracy of coordinate of the objects such that the robot can make decision based on it. This is a math intensive project.
Post a Comment