Posts tagged video

I like to make cool things

Today I had the last presentation of the first year of my Master’s degree in Multimedia. Therefore, for the next couple of months, I want to do nothing, hear nothing, read nothing in which acronyms are an essential part of communication. But still, I did learn quite a lot in the last ten months, and go the chance to work in cool stuff. The major highlights:

The first major work we were challenged to do was an e-book prototype. Me and my collegues Diana Carvalho, Eduardo Massa and Vitor Dioniso did A Cor, an e-book actually based on my documentary Words and Thoughts in RGB. We were asked to consider a futuristic e-book reader, therefore the e-book (which was actually put together the pretty common Adobe Acrobat) has some functionality still not found in current readers, such as color and the assumption there’s a touchscreen. You can actually download the e-book (it’s in portuguese, though — 10MB), but I advise you to actually open it with Acrobat Reader instead of the unbloated susbtitute you love, because the interesting things (draggable surfaces, pop-up, tiered text) only work there. We also intended this as a demonstration of a product concept I believe in: very cheap, short e-books about very specific subjects, the readable/interactive equivalent of the microdocumentaries.

Change Your Habits Today, done with Diana Carvalho and Eduardo Massa, was our answer when asked for an “interactive video” in which real and unreal footage are merged. Above you can see the non-interactive (here for HD), linearized video which I shot with my good friend Ana Margarida Carvalho. The final work, which is Heavy, has multiple layers of interactive Flash content my group colleagues added, so in a way this video doubles as its own ‘website’.

In the second semester I found myself another workgroup. Along with Juarez Braga, Manuel Almeida, Mariana Figueiredo and Marta Leal I did a mobile application called POC (don’t ask), which is an event-based social network. The kicker is that our app offers no way to chat or send messages to your buddies — you wanna talk with someone, you go attend the same event (in our prototype events consisted only of musical gigs). Of course, in order to prevent POC from being a robust stalker tool, you actually have many privacy choices. But we sure had some fun discussing the outrageous ways a social network could work.

Finally, today we had our last presentation. Keeping the same group that did POC, we were supposed to come up with an interactive installation, and so, after weeks of brainstorming, we came up with a display for the Faculty of Engineering Museum’s archives, which contain 19th century educational models in a closed climate-controlled room. In order for visitors to be able to ‘see’ the objects kept there, we made a touchscreen browser which controls a rear projection display with all the information about the selected item. Today’s presentation was about the concept though, and the display isn’t actually there (yet). We hadn’t an actual touchscreen controller (we had to do with a laptop and a mouse), and had to improvise the rear projection screen with sheets of paper. But still it went well as a demonstration of something that can be very easily and relatively cheaply be implemented in similar places.

Looking in retrospect, all this is hardly cutting-edge. I think that’s beside the point. There’s often a pressure to do with the most recent technology, that’s true, but I believe sometimes true innovation gets lost in that process. While I think a C64 Twitter client might be taking it a little too far, I think most innovation comes from the hacking of well-established technologies (just think of the low quality of games released for fresh new consoles), else it’s all eye candy and little use.

Oblong’s TAMPER system. AE Portal News has more information about this demonstration of the future of video editing. Now excuse me, I’m cleaning the drool off my keyboard.

Content-Preserving Warps for 3D Video Stabilization (via CDM): This is just an incredible presentation. Camera movement is one of the things that sets apart low budget from ‘professional’ filmmaking and it’s good to see the potential for yet another gap narrowing. Indiependents sometimes waste so much time and resources on it, it’ll be nice to let computers handle it so we can do some actual filmmaking.

Another useless trick

Actually, this is how I feel sometimes — perforated.

On with the Processing saga. One of the things I’m interested is in live video manipulation and EyeToy-like interaction with the computer. So the other day I went to buy a cheapo webcam so that I could do my experiments without risking an accidental punch at my Canon HD camera, besides keeping a cleaner desk. I entered an electronics store, picked up the ten euro camera that didn’t look like a total piece of shit (the images it generates, however, are the glorious crap-o-vision you expect — which is nice), and went to the cashier. Just as I was going to pay I realized that here was a nerdy, badly shaven creature about to buy a webcam. “Hm, quit the mid-90s prejudices and cybersex superstitions”, I thought. “People constantly buy webcams, right? This electronics store sells at least a dozen different models, so there’s as much demand for webcams as for, say, mice and keyboards. Besides, most laptops have incorporated webcams, and people have all kinds of uses for them… It’s just that I can’t think about anything else right now!” I paid for the camera, and the guy at the cash register winks at me. The fucker.

Anyway, I did buy my blob brightness and positioning sensor to get some work done. I’m trying to see if I can come up with a gesture-based interface for something we’re doing at my Master’s, but in the meantime I got to get my Processing confidence level high enough. As a first exercise, I did some eye candy: an application that converts brightness data to depth data. That’s not much, but I’m still learning my way around the JMyron computer vision library, which looks like the fastest shortcut to where I want to be. Since Processing web applets apparently can’t access people’s cameras (which is a good thing, I suppose), you can’t see my experiment online, instead you’ll have to download a compiled application to try it. Both the Windows and the Mac versions require, I believe, Java on your computer, and the latter version is untested, so I’d love to hear from you if it works. There are quite a few options I added to that program, press ‘H’ to see the instructions. Enjoy!

Download 3Dwebcam: Windows | MacOS X.

Matrix III, John Whitney 1972. (via CDM)

YouTube EDU

Lots and lots of documentaries and lectures.