Publications

Tom Yeh, John J. Lee, Trevor Darrell. Fast Concurrent Object Localization and Recognition. To appear, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Miami, Florida, June 2009.
Tom Yeh, John J. Lee, Trevor Darrell. Photo-based Question Answering. In Proceedings of the 2008 ACM International Conference on Multimedia, Vancouver, BC, Canada, October 2008.
John J. Lee. Efficient Object Recognition and Image Retrieval for Large-Scale Applications, M.Eng. Thesis, Cambridge, MA, May 2008.
John J. Lee. LIBPMK: A Pyramid Match Toolkit. MIT CSAIL Technical Report, MIT-CSAIL-TR-2008-017, Cambridge, MA, April 2008.
Tom Yeh, John J. Lee, Trevor Darrell. Scalable Classifiers for Internet Vision Tasks. In First IEEE Workshop on Internet Vision, in conjunction with Computer Vision and Pattern Recognition (CVPR), Anchorage, Alaska, June 2008.
Tom Yeh, John J. Lee, Trevor Darrell. Adaptive Vocabulary Forests for Dynamic Indexing and Category Learning. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), Rio de Janeiro, Brazil, October 2007.

Highlights of my time at MIT

My M.Eng. thesis, titled Efficient Object Recognition and Image Retrieval for Large-Scale Applications (PDF), won the Charles and Jennifer Johnson Thesis Award for Outstanding Computer Science M.Eng. Thesis (1st place).
My M.Eng. thesis later also won the Dimitris N. Chorafas Foundation Award. This is an international award given to about 20 or 30 people each year.
I won the Morris Joseph Levin Award for Outstanding Oral Thesis Presentation at the EECS Masterworks Symposium. I was one of four winners of this award. Here's my poster (PDF). See if you can find the typo.
I led a group of 5 in 6.087: Building Mobile Applications with Android. The class wasn't particularly interesting, and we didn't even get any Android hardware, but we got some decent exposure. Companies like Verizon, Nokia, Sprint, and some VCs came to see our final presentation. Here's the flyer (PDF) that we designed for them.
I took CMS.998: Videogame Theory and Analysis. I developed a theory about RTS gamers, their social interactions online, and its impact on the development of online communities. I gave a talk about my work and wrote a paper titled Action, Play, and Agency in RTS (PDF). Unfortunately due to other commitments I missed my chance to publish it.

I wrote mobile phone applications in 6.883: Pervasive Human-Centric Mobile Computing. We got Nokia hardware to play with and write S60 code for; we also played with some handheld Linux-based PDAs and indoor location services (crickets). My final project was a real-life Pong game where players had to physically run around to move their paddles.
I wrote a lot of music in 21M.341: Jazz Composition. It was a lot of fun. The musicians in our class formed an ad-hoc "band" and played everyone's final projects live in Killian Hall. Unfortunately I don't have a recording of it, but it sounded awesome.

The hardest class I've ever taken was 6.854: Advanced Algorithms. My group's final project improved the upper bound for a competitive algorithm for the general 2-server problem. An hour before we submitted the paper (PDF) we noticed a bug in one of our proofs, but we managed to patch it up and added a coversheet, explaining that the new bound was not as good but still an improvement. In any case, I'm not much of an algorithms person, but this class definitely whipped me into really good shape. I can safely say that the one point in time that I was in my smartest phase was right after I took this class. It has all been downhill after that.
During IAP, I did 6.270: Autonomous Robot Design Competition, where my team built a LEGO robot that had to move balls around. Here are some videos of our robot (ours is the one with the flag). Our loss in the 3rd round was due to faulty equipment on the organizers' part :-( (see the YouTube page for the 3rd video for an explanation).

I took 6.033: Computer Systems Engineering. There's not much to say here or any particularly good work to show off, but I put a lot of work into my design projects, which were titled ROFLBus: A Really Organized, Fast, and Low-cost Bus and FART: Fault-tolerant Archive Retrieval Technology. My TA's reaction to the 2nd paper was priceless. And so was our final presentation.

I took 8.03: Vibrations and Waves because someone told me it was an interesting class. And now I have a degree in physics.