|
||||||
|
NY Art Beat, a child of wildly-successful Tokyo Art Beat website, needs volunteers in order to get off the ground. If you’re interested in art, can find your way around HTML, and want to plug in to a great network of artsy people, check out what they’re looking for. At NYU’s ITP program, the Processing and Arduino programming environments are widely used. Since my main interest in the OLPC is to use it as a controller for projects, and since most of my projects are built using those two tools, I was very interested in getting them running on the OLPC. In addition, both Processing and Arduino are projects that have a lot in common with the OLPC project: they are focused on education, focus on making computing technology available to groups that previously would have found it difficult, are fully Open Sourced, and rely on a community-based approach to support and development.  Good news: it is strikingly easy (with caveats). The Arduino IDE runs under Sugar (the OLPC’s built-in user interface) and under XFCE (a more conventional Linux window manager). Processing runs fine (albeit slowly) under XFCE. However, it gets confused under Sugar, which doesn’t handle multiple windows well. Read past the break for instructions! I’ve continued to explore HDR (High Dynamic Range) imaging, trying out different tone mapping algorithms. The following image was processed with the Mantiuk algorithm, and gives a nice, illustrated look:  Compare it to the original image:  In imaging, dynamic range refers to the difference between the brightest and darkest values in a scene. Cameras are not able to capture as much dynamic range as our eyes, and most display technologies are not capable of representing them. I’ve been doing some experiments with HDR imaging and processing, because I think the process of converting real-life scenes into representations can shed some insight into what makes a painting look like a painting, instead of a photograph.  However, by taking a number of pictures of the same scene, but with different exposures, you can overlay them and reconstruct a picture where a lot more information is visible. This technique is explained here. I processed the same scene shown in the previous image, using 3 separate exposures, and the result is shown below:  Note how the view through the window is now much more detailed, while the room is much lighter. The scene looks more “natural” than the original camera exposure. It still looks very much like a photograph, not a painting. However, there is a fashion now, where HDR images are processed not to look natural, but to maintain as much detail as possible. This relies on using specific techniques for a process called “tone mapping“, basically the conversion from an (undisplayable) high dynamic range image to a lower dynamic range image. These images look very “painterly” and artificial. Check out the Flickr HDR group, for example. While there are a number of tools and techniques for achieving this exaggerated “HDR” look, I found one of the best to be free: QTPFSGUI. This piece of Open Source software handles two key tasks: aligning the different exposures, and “tone-mapping” the resulting HDR images so that they can be displayed on regular monitors. Here’s how my test image looks after being processed through this program:  The image looks much more like a painted illustration, than like a photograph. This is probably because the tone-mapping process is very much what a human painter or illustrator does. Rather than drawing everything “photographically”, an artist will tend to show more detail, more features. I’ve moved the information in this post to a dedicated page here. Just returned from Brazil, here are a couple of nice panoramic shots from Rio. The first is the famous Rocinha favela, the second is an overview of the Rio beaches taken from the Corcovado mountain. Click on the images to see the (huge) full resolution versions.   More photos from my trip here. I went to a party for Bug Labs, a New York based startup that is launching a modular portable hardware development platform. The core of the system is a very compact, powerful, Linux-based computer. It’s housed in a slick plastic case with four modular connectors that let you easily add modules. Right now these include a motion sensor, GPS, camera and LCD touchscreen, with more on the way.  It seems like a way of making development of electronic gadgets as easy and modular as software development. The whole project seems to be strongly informed by the Open Source movement, with most (all?) technical information readily available. Everything about the project seems very slick, very professionally developed. I especially like the software development environment: it is built as a plug-in for the Eclipse IDE, and includes a “Virtual BUG”, a full emulation of the Bug in software. Even the modules are emulated, though with some limitations: for example, the virtual camera will actually return images saved on your hard drive, the GPS module will return a pre-recorded data set. However, this should be enough for most actual software development. With that being said, there is no harm in seeing how somewhere like this nearshore software development company can assist you if you want to explore the specifics of this environment further. It doesn’t hurt to broaden your horizons after all.
Here is a screenshot of the BUG SDK in Eclipse showing the Virtual BUG:  This seems like a project with great potential. I’d like to see availability of more experiment-friendly modules, eg. a solderless breadboard, analog inputs, etc. Maybe even a way of integrating the Arduino microcontroller as a peripheral, since so much work has already been done on connecting that to various sensors. I recently participated in the One Laptop Per Child program that donates one laptop to the developing world for each one that you buy. Well, the little computers arrived today, ahead of schedule:  First impressions:
I will write more as I learn more. Another day, another weird van in the neighborhood…  Heading to school on Saturday I came across this:  Someone had turned a parking space into a tiny park, complete with official-looking sign:  I spotted a few other parks along the way, some of them being enjoyed by a variety of bemused people. It was all part of National Park(ing) Day, organized by the Trust for Public Land. Interesting concept, attempting to reclaim the streets for public use (and relaxation) instead of for vehicle storage. It reminded me a little of when a stretch of 3rd Ave. was covered in trees and grass for the filming of Will Smith’s “I Am Legend”. |
||||||
|
Copyright © 2024 Gian Pablo Villamil - All Rights Reserved Powered by WordPress & Atahualpa |
||||||
