I’m going to be working with Professor Baird this summer on a research project along with a few other people (that site has a description of his past work… it’s pretty ridiculous). He’s pretty excited about the project, and I am too, it should be really interesting (and hopefully fruitful). Here are a couple paragraphs explaining it from an email he wrote earlier in the week:

“Here’s one problem we might tackle, motivated by the pressing needs of intelligence analysts: given a huge number of images (B&W, color, grey) some of which are of documents (nobody knows which ones), detect the presence of text (handwriting & machine-print), the reading direction of the text, the approximate number of characters/words; also detect signatures, logos, tabular data, scribbles, dirt, etc.”

“We’ll strive to make very few if any assumptions that constrain the images and document types (e.g. they can be in any language), and we’ll try to build classifiers that are automatically trainable
using massive synthesized data (*no* handcrafted special cases), and run extremely fast (e.g at “I/O rates,” as fast as the images can be read….!).”

Hopefully this will be successful and publishable, or at the very least help me figure out what I want to do for a thesis and what I want to do after Lehigh.

When I went in to meet with him today, instead of asking for a resume or anything (he doesn’t really know me much outside of class), he just looked me up online. This made me think that I should probably update my lehigh.edu page to reflect my more recent work and interests. I’ll try to get to that before the summer.

Today I donated $50 to the MS (multiple sclerosis) walk in Pittsburgh. Granger linked to his friend Alisa’s entry since she is raising money for it. My uncle has MS and happens to live in Pittsburgh, so it was especially important to me. I recommend reading Alisa’s story and check out the links in the entry.