Marynel Vázquez – (r)evolve
This work is an attempt to generate interesting graphics from human input. I used OpenFrameworks + ofxKinect + ofxOpenCv2 (a simple add-on I created to interface with OpenCV2).
From the depth image provided by the Kinect:
1. edges are detected (Canny edge detector)
2. lines are fitted to the edges (Hough transform)
From the color image:
3. Colors are taken at the endpoints of the lines, and painting is done by interpolating these values
The result of about 10 frames is stored, and then displayed. This allows to perceive some motion evolution.
This line of projections was then multiplied, and the width of the edges were allowed to change according to how close stuff was to the Kinect. By moving the camera around the model, one can generate images like the following:
The video below shows how shapes can change,
Interesting additions to this project would be a flow field and music! In the former case, the generated lines could be forces that influence the motion of physical particles in the 3D space.
Hi Marynel – nice exploration. Here are the notes from the crit. -GL.
————————–
This could be a cool performative interface. You just said it. Music.
Seems a little too jittery—no, flickery. Jittery is good. The flicker is disorienting.
Nice that you figured out how to make an addon. That process is a pain in the ass.
This iss cool! Nice human interaction with the generation. I kind of which it would save some of the lines, so you could generate something with a few poses or something, instead of it being lost right away.
It would be great if one hand controls what is drawn and the other controls rotation/speed/etc.
It’s difficult to understand how the visuals match the image, but it’s a really smooth display of technology that brings together a lot of difficult concepts.
It really became interesting once you showed it with the laptop case, I was not understanding that the lines were the actual edge objects, its great, but with more polish it will be really cool. The rotation seems a little random, I think a little more thought in how the lines are revolved around and how they array themselves out would make it even better.
I agree that it was really cool with the laptop case / other shapes … If you could save it out as a 3D model, it would be an incredible modler for architectural applications.
This is really beautiful. I would love to so prints.
Bringing music into it would be really cool. The visualization has great rhythm and reminds me of a large choreographed dance
very nice looking and abstract. the movement it very energetic.
It’d be cool if you added music. Nice visual.
I feel like this needs some careful aesthetic editing — in particular, the length of the lines varies wildly from impossibly short to infinite. Perhaps take the log of the line length?
I also feel like this will be improved by more thinking about its context — an application (even a highly fun/gamelike/artistic one) for which these patterns are used.
May be faster to compute local orientation of blobs with moments, than applying Hough transform to canny edges. Hough is very slow.
This looks really fun to play with. I like the surprise of what kind of pattern you will get.
Kind of like a 3d kaleidoscope. I can see how this would be engaging to play around with.
Definitely looks cool (pencil sketches!). It’ll be great once the perfect context comes by.