Haar Classifier Training Program

If you haven't used Processing to power through some small application that you need, I highly recommend it. (I just now remembered I still haven't posted about my poster in the conference, and given that it's officially published, I can officially talk about at least the extent to what we've done so far. Next post will be on that.) In any case, as part of the research, we're now switching to an OpenCV implementation, which means training a Haar Classifier. Now this is a task that is...a pain. It involves ideally thousands of images, and if you want the best accuracy, you should hand-annotate where your object is in the images.

To avoid going crazy and needing to destroy everything in the lab, I wrote a short app in processing to create the Haar file. I'll probably post a link to the code after I've completed the training on our images and verified that it doesn't start eating memory like no-one's business for large samples, since I had tried so very hard to avoid writing this app myself. Still, using Processing, it took all of 30 minutes to produce something that will be useful. I could've spent more time with the interface, and if I decide to release it, I very well may modify the code to have a nicer I/F and support multiple occurrences in the same image, but for now, this little app is sufficient for my purposes.

For the record, there's something utterly satisfying about completing a program in under an hour. :)

UbiComp 2010: Paper of MY Conference

So while this paper was not nominated for best paper, and should not have been, it did quite a bit for me personally in terms of considering structure for my own future thesis.

Andrea Grimes, Vasudhara Kantroo and Rebecca E. Grinter (2010). Let's Play! Mobile Health Games for Adults. Proc. of UBICOMP 2010. (Link is to Ms.Grimes website, and you can find the full paper on the publications page.)

Let me sum up the gist of the idea, rather than going into the details of the paper, as there are some fundamental problems with the study. The gist is that a casual game might be better suited for games that aim to improve behavior. The behavior here was based on healthy eating, and the research targeted African American individuals because of the high incidence of diabetes among that population. The game was designed with food items that such folks would find in their neighborhoods, including soul food type items, sandwiches, etc.

The critical part for me was the introduction of the casual game premise. I had been structuring my thesis idea around epistemic games, thinking only in terms of long story arc type things, and listening to this proposal has given me an alternate idea. I'm definitely going to need to start with an exploratory study to examine which of those two paths is going to be the better choice for me, I think. Additionally, discussing the shape of the study and what might have been a better question for the study format has helped me think about how to structure my own studies to avoid those types of critiques.

I'm likely going to blog about a few more of the papers that were interesting or that I discussed in a critical fashion with Leilah as well.

Academics: Very Smart, Very Dumb

I have one major complaint about UbiComp 2010 as far as comfort goes. The seatting arrangement during the presentations is a typical lecture hall, but one with absolutely no space to move if someone is sitting in a chair. For some reason, every single person will opt to sit on the very end of the row. This means that every person who is not in the lecture hall before the presentation begins has to squeeze past those folks who sat down first. This is particularly difficult as some folks refuse to stand up to let people pass.

This is particularly humiliating in my case. I'm not a small woman by most standards, and it. is. not. comfortable. for me to move myself and my laptop pass you. You're forcing me to choose to show you either my rear or my front in a small area. It is awkward for me. I can't imagine you're enjoying it that much, given your reluctance to even stand up. When I say "Excuse me," it isn't a request for you to turn your knees to the side, which adds maybe three inches to the first three that were there. I'm simply not six inches wide.

So please, colleagues. If you're organizing a conference, do choose a venue that it is accessible for various body types. If you're at a conference that doesn't have wide aisles, either sit near the center of the row, or be prepared to stand up when people need to get by.

A further update about UbiComp as far as interesting papers went will be posted either later today or tomorrow.

Software Review: Mendeley

Finding a resource management software ended up being really easy. I searched cross-plastform because I work on a Mac in the lab, a PC at home, and up until today, my laptop was running ubuntu. (Since ASUS is no longer providing drivers for linux, and I really wanted to be able to skype with Braden while in Denmark, I'm installing Windows on it, at least temporarily.) Mendeley is absolutely fantastic. It scans pdfs to try and autofill citation information, while allowing you to sync files or information across computers. I'm so into this software right now. I have't encountered any bugs that I want gone or features that I'm desperately missing.

Mendeley exports to EndNote if you happen to have access to commercial software, as well as exporting to BibTex. I'm digging it so far. I'd chat it up a bit more, but I need to put together my personal-item purse/bag/whatevs for the flight to Denmark. :)

Catching Up/Getting Ahead

It's been a while because I haven't yet gotten through my thick skull that blogging more frequently might actually assist me in keeping my thoughts straight about the various projects I'm working on. This semester started all of two weeks ago, and things haven't really started to kick in full gear yet as far as academics are concerned, though I've been out of town for the last two weekends and have two more social engagements on the next two weekends, and then get to take a plane to my first ever real-computer-science-conference.

What? I never posted here about that? Right, well the poster paper on Paper-to-Parameters was accepted to UbiComp. On September 25th, I take my first international flight using my first passport(which cost me way too much money), to go to the first conference where my first paper ever will be published. I've spent a lot of time in the last week working on the poster, and working considerably harder since realizing that this may very well be a conference I'll want to be taken seriously at when I'm working the angles on my thesis.

Question to be asked at the department sometime soon: can we publish partial parts of our PhD research in a similar way to how the work we're doing right now is being published as a WiP in a couple of conference before the final paper is published? I really don't know these things at UIC yet, and I have a feeling it could be critical.

I'm planning on applying to the NSFGRF program this year. Getting accepted means some $30,000 a year stipend, along with an allotted tuition waiver, as well as travel assistance. It would also mean free time to actually do my research without having to work for someone else. The other nice thing about the NSFGRF is that it basically sets me up to get background work done for my thesis and to put some really intense thought into what I want to do and how I would do it. It's basically a mini-grant proposal. I'm actually really excited to start working on it, if only I could start budgeting my time better.

A couple of things I'm going to try and do: find a reference management software that I like, and then leave the references in my dropbox folder so I can access them from all three computers. Write up short reviews here again for papers that I've read, because well, I should be doing that anyway as I finish the papers to avoid re-reading too deeply the ones that are uninteresting/disconnected from what I'm doing. Discuss with Leilah what is appropriate to be spoken of with the internet before publication so that I can figure out just how much to reveal here about what I'm currently working on. Some open source software reviews for various toolkits I've been having to use.

For now, that's catch up time. I'm working on putting together a study group for the Foundations exam amongst my fellow graduate students, so that may be updated here as well.

CRA-W Grad Cohort

This weekend, I took a break from the last-week-of-school panic to attend a conference in Seattle for women in computer science. It's actually not so much a conference as let's all get together, eat free food, and talk about how to get through graduate school. And for the most part, I learned quite a bit and enjoyed myself quite thoroughly and actually made some new friends.

There were certain parts that concerned me. The two sessions that one wouldn't have seen at a similar event for any gendered cs grad student were...pretty gender biased. In a bad way. The suggested reactions to possible situations made me uncomfortable. They were all very passive reactions, which only leads to the persistence of the attitude that it's okay to say that someone only got their position because of their gender, or to interrupt a female speaker just because they're female. I do not believe in simply sitting back if I'm having a conversation with someone and someone else butts in and tries to end my conversation preemptively. I am far more likely to inject myself into the "new" conversation, politely. Restraint is something I should learn.

But there were constant references to "having a cry" when a rejection letter was received. There were the 'jokes' about shopping. If we're talking about activities that are relaxing, why are they all so stereotypically gendered activities? It...perturbed me. For all of the information I received, the attitudes I encountered at CRA-W were not really...about being equal. They were about establishing our femininity within being a computer scientist and somehow trying to be normal.

I have no desire to be normal. Not a normal woman, whatever that means, not a normal computer scientist, not a normal anything. Will I return to CRA-W next year if they fund me? Yes, I likely will. The opportunity to network with women in computing research is much too large to ignore. Will I perhaps be skipping the sessions on "Building Self Confidence" and "Being a Woman in Computing Technology" ...no, because I think it's critical that someone stands up and says that we should not let even "unintentional" sexism stand.

On a somewhat related note, if ever you have to be stranded in an airport for two hours, SeaTac is a great one for it. Comfortable seats, true free wi-fi, and outlets galore.

Qualifier Results & Summer Plans

Qualifiers Scores:

Foundations: No Go. Retry in September. We received the exams back, and I really need to study algorithms in depth, as well as some dynamic programming techniques. Whoops.

Human Centered Computing : Conditional Pass. I have to take Artificial Intelligence this Fall, which is cool. My AI course was not a practical course, instead being largely a theoretical approach. UIC emphasizes practice so much that I'm actually mildly excited. I did rock the HCI portion of the HCC exam, and that delighted me.

Systems: Pass. I find it peculiar that the only test I passed absolutely on the first try is the only one I didn't explicitly study for, but I'll take what I can get. Primarily my success here was due to taking Networking the prior semester.

While I attempted to apply for internships, I was largely unsuccessful. For some reason, places that are looking for research interns actually desire research experience. Fancy that. Luckily for me, I've been talking at length with Dr.Leilah Lyons about a project that only recently was given a small budget so that my summer will be partially funded, which is a delightful development that makes my summer much less stressful. The project?

The project involves two subjects that I'm aware of and one that I've studied with just a bit of depth. Essentially, it's an augmented reality project that will take AR input for a system, and then run a simulation for that system, based on the input. I actually question the extent to which this is Augmented Reality to some extent, since it's being used as input only, and not active input. Once the images are captured, they are not then acknowledged again unless another capture is requested.

The semester is coming to a close and Kenchi Games is taking up quite a bit of my time. Reach has been an interesting project with many, many problems that we're working on, a day at a time. Hopefully, we'll be making another video capture this week to post on youTube. The original intent had been to do that far more often than we have, but the amount of progress between weeks has been less than desired.

Additionally, for User Interface Design, we're now working on small projects in Android. This slow build up will hopefully lead to a good understanding of the API so that I can build small applications over the summer and increase my portfolio of work. (Speaking of portfolio, another high priority item on the list is definitely completing a portfolio website by the end of summer, since the school year seems to not include much real time.)