I'm Made of Gingerbread, So...

Watching Knocked Up today for the second time reminded me how hilarious Kristen Wiig is. While Katharine Heigl and Seth Rogan and Paul Rudd were all completely brilliant and perfect for their parts, my greatest laugh-out-loud moments definitely came from the two, short Jill/Alison scenes:

Somehow Wiig manages to plant the bratty high school girl into the sophisticated business woman. Basically, it's Mean Girls aged by about 10 years. Absolutely love it.

Not too far of a departure from that character is Penelope Morris, a recurring character from SNL who is best described as a "habitual braggart". If you watch this video and don't immediately adore Kristen Wiig (and I mean Kristen Wiig, not Penelope Morris), I will give you a quarter.

(If you can't see the video above, go to this link).

My Fireblog: Trip to Paris. I mean, Perris.

Sign on San Diego's "fire blog" has been the most up-to-date source of news, including info on new areas that have been evacuated and/or caught fire. But for those of us without land lines, and without knowledge of how to register cell phone numbers with the Reverse 911 system until like two days into the whole fiasco, this website also served as our freaking lifeline. It's kind of sad to have to rely on the internet to tell you when you might have to run for your life... Because of this, I gave it a permanent spot on my desktop and refreshed it more frequently than I did my undergrad transcript when grades were supposed to come out. And that's saying a lot.

Early Sunday afternoon, I began to smell smoke. It was hot and dry and eerily silent despite strong winds; the palm trees outside my bedroom window made foreboding crackling sounds as it rustled. By Monday morning, the air had a dead yellow tint to it, and I could see ashes collecting at the bottom of our swimming pool. The sun was a bright red ball in the sky, and at dusk it produced a magnificent, almost synthetic-like sun set with gorgeous shades of bold orange, reds, and pinks. Too bad it would be at the cost of so many homes and exasperated firefighters.

Sometime on Monday night, we came to this realization. There were four fires bordering the north, east, and south sides of us, and two of the three major highways out of the city had been closed. The powerful Santa Ana winds were not expected to die down until Thursday. It seemed like a new county was being given the "go" every hour or so, and Del Mar and Solana Beach -- only about 15 miles north of us -- had just been evacuated all the way out to the coast. Two fires raged within 15 miles of Qualcomm Stadium, the evacuation shelter for our area, and it was already a fifth of the way full, having not evacuated any of central San Diego. The air was hardly breathable and it was hot and stuffy inside the condo with all windows and doors closed.

All of this makes it perfectly logical for me, after receiving a call from my energized-and-adventure-prone friend at 2 a.m., to jump the gun, pile into two cars with five other people, and "book the hell out of the city" on the last open freeway. Yeaaa. And that is all I am going to say about *that* little fiasco...

Two discoveries that came out of this whole experience, both of them (unsurprisingly) singing Google's praise:

1. GOOG-411 is actually really useful when fleeing a fire into an unknown town. We arrived randomly in Perris, CA at around 5:30 a.m., because we were too tired to drive any further, and were far enough away to have clean air and a peace of mind. When the fancy, built-in GPS system on our car failed to give any valid hotel numbers, GOOG-411 saved our butts. It is the sole reason for me being asleep in a hotel room by 6 a.m. Later, we used it to find restaurants, starbucks, video rental stores, movie theatres, etc.
   While Verizon Wireless (and other similar carriers) offer 411 services that charge by the call and make you talk to a real person, GOOG-411 is totally free and completely automated through speech recognition. That means, you don't feel bad when you hang up mid-call. It's 'text message' option (which texts you all info including address and number) is also extremely useful, especially when hotel personnel have enunciation problems...

2. The open source aspect of Google maps has played a huge role in monitoring the status of the fire. Sites like latimes.com, kpbs.org, and signonsandiego.com were able to offer interactive Google maps that pinpointed not only areas of evacuation and fire, but also shelters, road closures, smoke advisories, homes destroyed, etc.
   Its inherent feature for being able to tie information to location and then search through listings also proved indispensable for this purpose. For example, details about the number of homes destroyed, acres burned, firefighters on the job, etc. were tied to each fire (click image to enlarge).
   
   It is a powerful interface for relaying geographically significant information and I think very useful natural disaster related applications can be built by combining this interface with the "wisdom of the crowd". For example, it would've been helpful if the average person could contribute reports on what they see/hear/smell of the fire their location on the map. They could even discuss planned evacuation routes, report on traffice in the area, their way of getting contacted for evacuation, etc. (replacing the several active forums that have sprung up over the last couple of days).
   Shelters/firefighters themselves could contribute to information and numbers, providing more abundant, reliable, and up-to-date information. Information (data-mining? or just good old manual input) from blogs, news reports, or radio broadcasts could also be stripped and added on. Perhaps a digg-like rating system could be put in effect that would allow most popular/reliable reports to show up most visibly (i.e. on the least zoomed in version of the map). If monitored correctly, information would flow bottom-up instead of top-down -- something that has been proven successful more than once.

For the record, I was pretty impressed at how smoothly the evacuations seemed to progress. If only they had broadcast info earlier about how poor grad students who don't own land lines can sign up for Reverse 911. Is this some kind of occupational-profiling?

SoCal On Fire!

This was one of the hundreds of headlines I've seen in the last two days. On the many interactive google maps that have been put out by various news stations, it literally looks like Southern California is on fire, with over 15 different fires raging in parallel:

Some pretty crazy photos have been collected here.

While people in La Jolla (where I live) and most of central San Diego have been untouched for the most part, others were not as fortunate. Hundreds of people have lost homes, and hundreds of thousands have been evacuated. If you are from the San Diego area, UCSD has put out a page that describes how you might be able to help. Otherwise, cash donations can be made to the American Red Cross here.

Ode to Audio

Every quarter is different, but there's always this very distinctive turning point about four weeks in, when you go from "i'm so excited about this course I'm taking!" to "JUST GET ME OUT OF HERE ALIVE." in the span of like 3 days. I think this might be day 1 of 3, so I thought I'd be nice and blog before I joined the dark side. Also, for the record, professors who give out easy introductory assignments and follow it with a second, massively harder assignment after the drop deadline, are evil.

So, one of the courses that I am really excited about this quarter (!) is Serge's object recognition course. Not that I am all of the sudden interested in computer vision... but there has actually been some ongoing research that tries to apply methods from (visual) object recognition to auditory object recognition. Briefly: object recognition determines if some specific object appears in a certain image; auditory object recognition determines whether some specific auditory "event" (i.e. spoken word, musical melody, sound of a door slam) has occured in a given audio clip, usually with a noisy environment.

For me, auditory anything was pretty much a non-existant concept to me before coming to UCSD, especially since most all of UW's speech/audio people resided in the EE dept, and few mingled with us. In general, it is also true that the computer vision field has flourished in the last two decades while its auditory counter part has not. Is there a bias for one sensory system over the other? Maybe. From my undergraduate perspective, any mention of the term "recognition" automatically applied to the computer vision field -- robots/machines that could "see" the way that humans do. It was always about hardware dogs that could play soccer, or search engines that could find pictures based on image content, or systems that could recognize hand gestures... the list goes on.

So when I was first introduced to the idea of computer audition -- engineering intelligence that can "hear" the way that humans do, I had one of those great "O_o" moments. Which is precisely what led me to UCSD in the first place.

I think my initial interest in computer audition was partly sentimental. Like many people, I'm crazy about music. Specifically, classical. I don't know much about signal processing or even music theory (sadly, even after many years of piano playing), but the simple fact that my auditory system has allowed me to enjoy so much music over the years seems to warrant at least some sort of return....as weird as that sounds.

Apart from being just really interesting, the field opens up a whole new world of possible technologies; many that can complement those that are being/have been developed in computer vision. If an intelligent machine has eyes, why not ears? These applications will have great educational, musical, informational, military, and medical impact; all of them inspired by aural abilities that most humans take completely for granted everyday. For example, we are currently thinking about possible technologies that may help the deaf or hard-of-hearing community.

The following is a set-up that gets right to the "heart" of what we wish to be able to do computationally. This is written by Albert Bregman in the classic book, Auditory Scene Analysis:

Imagine that you are on the edge of a lake and a friend challenges you to play a game. The game is this: Your friend digs two narrow channels up from the side of the lake. Each is a few feet long and a few inches wide and they are spaced a few feet apart. Halfway up each one, your friend stretched a handkerchief and fastens it to the sides of the channel. As waves reach the side of the lake, they travel up the channels and cause the two handkerchiefs to go into motion. You are allowed to look at the handkerchiefs and from their motions to answer a series of questions: How many boats are there on the lake and where are they? Which is the most powerful one? Which one is closer? Is the wind blowing? Has any large object been dropped suddenly into the lake?

Solving this problem seems impossible, but it is a strict analogy to the problem faced by our auditory systems. The lake represents the lake of air that surrounds us. The two channels are our two ear canals, and the handkerchiefs are our ear drums. The only information that the auditory system has available to it, or ever will have, is the vibrations of these two ear drums. Yet is seems to be able to answer questions very like the ones that were asked by the side of the lake: How many people are talking? Which one is louder, or closer? Is there a machine humming in the background? We are not surprised when our sense of hearing succeeds in answering these questions any more than we are when our eye, looking at the handkerchiefs, fails.