Thoughts on the September 09 New York Tech Meetup

Last night I attended my first session of the New York Tech Meetup, and it seems I’ve found exactly what I’ve been looking for in the city when it comes to a vibrant tech community. I recently moved to New York from Western Massachusetts to pursue a career writing about technology and film–after several years of IT work and dreaming about making the jump to writing full time. I had already gotten my taste of the film blogging world after covering Comic-Con and several press screenings for Slashfilm, but I had yet to get a full sense of what being in New York means for my tech interests until last night.

The first thing I noticed during the evening’s introduction (by NYTM organizer Nate Westheimer) was that New York techies seem to have a major complex towards the Silicon Valley and Boston tech communities. There’s a clear cultural difference between NY and the Valley, but as someone who spent their entire life halfway between New York and Boston (I grew up in Hartford, then went to college and worked for four years in Western Mass.), it never really occurred to me how the NY tech community was different than Boston. Westheimer mentioned one key difference (which he also wrote about on his blog)–that the NY community fails at connecting the university and commercial spaces.

Given the tech titans we’ve seen come out of MIT and Harvard, I think he may have a point. Culturally, Boston tends to focus on academia and intellectualism, so it’s no surprise that their higher educational institutions aren’t ignored when it comes to commercial tech ventures.  Westheimer went on to explain that the scheduled demos (two from Columbia, and two  from NYU) were examples of the innovation coming out of the university space, and that the commercial sector should take note.

Here’s a rundown of those demos:

Green Dot (NYU)

On their site, the Green Dot folks describe their work as “a research project that investigates motion capture, pattern recognition, and ‘Intrinsic Biometrics’ techniques to detect and analyze human movement signatures in video.” In short, their technology resembles the motion capturing we see today (wherein an actor’s movements are recorded via dots placed on their body), except without the need to wear anything. After a video feed is processed by their technology, the end result is a video with green dots tracking the movement in the source. Again, this looks very similar to footage you see from typical motion capturing techniques today.

They demoed some video footage of famous figures and pointed out some interesting factoids based on their visual signatures. The project is government funded, and you can definitely foresee some military uses of this technology once it’s perfected. They describe their current focus as “the analysis of national and international public figures while they are giving speeches, with future plans to investigate many other domains.” Read into that what you will.

On the consumer and professional levels, the technology could be really useful for bringing motion capturing to the masses. Imagine taking a video of your cat walking, and then applying that motion to a 3D figure.  If you’re an iPhone 3G or 3GS user and want to try it out for yourself, you can download their application from the app store, just search for “Dot Show Ian Spiro”.

CuZero (Columbia)

CuZero is an interesting attempt at changing the way we search for images. The presenter points out that image searching in its current fashion simply doesn’t work very well (anyone who has used the Google Images search for any length of time can attest to that). For the most part, image searching today relies on file names, the “alt” tag (which can be placed alongside the images on a website, but usually isn’t), and text surrounding the picture.

CuZero turns this paradigm on its head by allowing you to bring in text and image queries into a grid, and prioritize the different elements of your search in real time. You can also further filter your search results via a similar interactive method. It’s honestly hard to explain in words, so head over to their website and check out a video demo.

I got a very Minority Report vibe from the way they were able to interact with their queries, and the way their system also recognized buildings and other elements from pictures was also fascinating. This research certainly has the potential to change the way we search for images in the future.

Musically Intelligent Machines (Columbia)

Musically Intelligent Machines is a project that aims to automatically classify music–much in the same way Pandora works, except without the need for human intervention. The developers crawled through 15,000 mp3s to demo the service, and you can play with it on their site. The presenter mentioned that he was surprised the software could determine what constituted “British” music. Apparently it’s smart enough to determine British accents, and possibly even British styles of music.

Overall, the software works as advertised, and there is certainly a market for this sort of thing. I’d be interested to see how it would compete on a commercial scale against the likes of Last.FM and Pandora.

Teaching Robots to See (NYU)

At these sorts of events, there has to be at least one project that portends the impending robot uprising, and this project’s goal of creating robots that can see certainly brings Skynet to mind. NYU professor Yann LeCun demonstrated the research by showing how their software can recognize objects (he used a toy soldier, which was recognized as human, and a toy airplane that was recognized as an airplane). He also showed how they’ve trained robots to move autonomously and avoid objects using a dual camera system for eyes.

His work actually comes out of the same place that Green Dot calls home, the Vision Learning Graphics group at NYU. With their technologies combined, we could easily see some very important future developments emerging.

---

There were also some short demos by Donors Choose, which looks like a great way to support teachers across the country with donations of all sorts, and My Teacher My Hero, a site which focuses on stories of exceptional teachers. My Teacher My Hero will be launching on Sept. 8, and they’ve already lined up some famous voices to talk about their favorite teachers. As with everything these days, the site will also have a social component to allow you to share your own teacher stories.

Dan Bricklin was also interviewed by the great NY blogger Anil Dash. Bricklin was both hilarious and insightful, and I now feel compelled to grab his new book at some point. There’s something about tech history stories that I find completely fascinating, and as creator of the spreadsheet (among many contributions), he must certainly have some good stories to tell.

Check out Dan’s interview, along with the rest of the event, at the NYTM video page.

More than 1,000 words later, I realize now that I could write about this Meetup for several more pages. I take that as a good sign. The NYTM makes me excited to be a techie in New York, and I’m proud to be the third person to RSVP for October’s event. If you’d like to join, simply sign up via their Meetup page. I would love to have some Meetup buddies!