Archive for the 'animation' Category

Data doodles

Saturday, August 13th, 2011

About a month ago, I made a new backup of the data from tinyDoodle. It is available as a text file consisting of 31.2 megabytes of integer coordinates of 2d points that are put together as a very long sequence of line segments. It’s formatted in JSON in a straightforward way. It doesn’t matter to me how silly this application sounds, there is something I still find incredibly compelling about the ability of computers to capture drawing gestures as sets of numbers that can be performed as drawing gestures that are sets of numbers. I think this drawing-to-number quasi-biyection is priceless.

I was recently talking about how different interaction models determine differences in communication, and how interesting it is for me to look at scenarios where a group of humans is restricted to use non-conventional channels to communicate with each other. Like putting two persons in a room and have them play a game where all they can do is make drawings to each other. Blackboard, paper, whiteboard, it doesn’t matter. Their communication will not be very efficient this way, but they will get very creative at drawing, and maybe come across some ideas that they would have never explored any other way.

More recently, Buzamoto launched a cool iPad app called Pendipity that offers a similar functionality to tinyDoodle, only better. It features a more advanced, yet very simple, drawing interface, and it implements a seamless chatting experience using a Node.js server. In terms of space, the difference between both systems is clear. When someone initiates a shared Pendipity session, the system will look for another available user to create a drawing team of two, and TinyDoodle is an open space where anybody can access the same drawing at any given time. So tinyDoodle is like a public blackboard, and Pendipity is like a shared notebook where every visitor is paired with someone else to draw on a single page of the notebook at a time. In Pendipity, a different session means a different drawing. In tinyDoodle, there will always be the same single drawing, around thirty something mb long at this point. The drawing is so dense, you actually have to watch it in chunks to make sense of it.

The following image is a collaboration Buza and I made on Pendipity. We didn’t find out we were drawing together until later, when we talked about it by chance. The idea of collaborating with somebody close to you without knowing who they are is bizarre, to say the least.

Geometry is back

Monday, July 25th, 2011

This weekend, I have been swimming inside a projection of the 120-cell courtesy of Jenn3D. The tetrahedrons stand for vertices. Jenn3D looks great. I downloaded the source code but I couldn’t understand most of it. At least I got it to compile.

Stop Motion experiment

Saturday, August 14th, 2010

A tripod on a carpet is pulled forward from under a table to deliver the perfect DIY dolly motion. The rest was shot frame by frame. Work in progress for the Cambridge based film collective FANSO.

Between 2D and 3D

Sunday, August 8th, 2010

This is update from a previous note. A few months ago I modeled a few cartoon characters using an experimental modeling application developed by Alec Rivers at CSAIL. Working with it is actually a hybrid process between drawing and modeling. After drawing a few views of a cartoon character from a few basic two dimensional shapes—front, side and top for example—the software tries its best generate all other views required to look at the character from any p.o.v. in three dimensions. An iterative process lets you refine the views that don’t look right, rearranging and deforming the original shapes, until you build a two dimensional character that can be looked in three dimensions from any angle. Hence the name of the project: 2.5D. I believe using this software can be significantly less confusing than my explanation. Alec and his collaborators are definitely more clear in the paper that was featured in Siggraph this Summer. If you visit Alec’s project webpage you can actually download the software and play with the models I made—or make your own—provided that you can run Windows 7 or Vista in your machine.

The character featured in the picture combines features from Disney’s Stitch and the little green aliens from the Toy Story series.

I am not sure if a version of this technique will ever become an industry standard. It all depends on how much smarter computers will become in the future, but it’s a good reminder that the creation of new digital tools is an open door to new forms of expression, even within the constraints of traditional forms like cartoon animation.


Thursday, December 10th, 2009

This fall I worked on a top secret CSAIL project, modeling toon characters with an experimental system that I can’t talk about until it goes public. This job has reminded me how much I love cartoons in general, and how I should be doing more of those, and less of other things.

Cartoons sit halfway between realism and typography, still kind of faithful to some aspects of realism, but conceding a lot to symbolic representation. It’s not that cartoons can’t represent things faithfully, cartoons choose not to do so in order to communicate things better.

Cartoon shapes and environments can’t be fully defined in terms of geometric systems and mathematical modeling, forcing the intervention of the human component that is the essence of many deep cognitive questions. Cartoons are Gestalt at its best, and they are also fun as hell.

IAP – Teaching Animation

Wednesday, February 4th, 2009

Early this year I was offered the opportunity to teach animation over IAP by the MIT Student Art Association. IAP, or Independent Activities Period, is a special four week term at MIT that runs from the first week of January until the end of the month. The class would consist of three weekly sessions of three hours each, and I was required to start teaching right away.

Originally, I was asked to teach a software based 3D computer animation class, but I preferred to forget about computer software and approach animation from a more general perspective. I think learning animation is more exciting -and useful- than learning how to use a computer program. The most sophisticated animation software in history doesn’t help to become a good animator, but a basic set of animation skills can easily be applied across a broad number of mediums.

My first challenge was to find a way to remain entertaining for 3 whole hours. I summoned my favorite moments from Frank Espinosa’s teaching—here at MIT—and from when I studied animation at VFS, including a copy of the classic Illusion of Life by Ollie Johnston and Frank Thomas.

Having talked about some basics and history of animation over the first session, I needed to outline some kind of program that could be run as a workshop over the remaining couple of sessions. My purpose was to help the students produce simple animations from scratch to completion over a 3 hour interval, and use the results to start a discussion on storytelling, timing, and animation.

After being introduced to Michel Gondry’s Be Kind Rewind Protocol, where he describes a program for people to produce amateur movies over an amount of time similar to what I had in mind, I adapted his approach to design a number of animation recipes, hoping to help the students focus on a range simple enough to conceive a situation and animate it in a short amount of time.

In the end, Computers and digital cameras played an important role in facilitating a quick way to test and screen the animations, but the use of editing or animation software was avoided. I wanted motion to be controlled by adding and removing frames by hand, so that none of the thinking during the animation process could be delegated to the computer.

The following pictures are frames from a Stop-Motion animation created last Sunday by Vvva and Lezno PlaK using a tatami mat, a broomstick, a hand cut banana peel, a hand crafted paper robot, and a Motorola Razr cellphone camera to capture each frame. The recipe they followed required them to create a biped character, and think about a situation where the character would walk until an external force stopped it.