Action Paintings – Jeremy Rotsztain
[vimeo=https://vimeo.com/24178732 width=”600px” height=”400px”]
This is one of the series of animated digital paintings made by Jeremy. The idea is to show the actions, sounds and animation sequences from movies (Terminator, Bourne Series, Fight Club) in the form of paint splatters on the canvas.The clip was created using custom rotoscoping software that transforms visual artifacts from aggressive cinematic gestures into vector shapes that look like paint as it splatters on a canvas. I like how the artist is transforming sound sequences into equivalent painting strokes, more or less like kinetic typography, but the emotions are shown through paint splatters.

Node Garden

Chemical Morphogenesis, Kaustuv DeBiswas & Alex Tsamis

This project is a form generator using diffusion-reaction, inspired by Alan Turing’s 1952 paper on morphogenesis.

The following citation from Turing’s paper (cited by the designers) describes the dynamic system:

“A system of chemical substances, called morphogens, reacting together and diffusing through a tissue, is adequate to account for the main phenomena of morphogenesis. Such a system, although it may be originally quite homogeneous, may later develop a pattern or structure due to an instability of the homogeneous equilibrium, which is triggered off by random disturbances.” – Alan Turing, 1952

Biomimetic butterflies, The Barbarian Group, 2006

Project’s webpage:

http://mcleodbutterflies.com/

Using a series of generative algorithms for the structure and pattern of the wings, seven unique “species” of butterflies were designed. Some of the generative methods used were voronoi algorithms, a simulation that shows the direction of the fluid movement as curved lines, cracking and circle packing algorithms as well as intersection lines algorithms.

The produced designs were used to produce an interactive installation of mechanical butterflies that were responding to the observers by flapping their wings!! The mechanical butterflies were composed by a laser-cut model , stepper motors, rare earth magnets, circuitry and a Mac Mini driving the installation using input from a video camera.

Michael Hansmeyer, Subdivided Columns

This project involves the design of a new column series based on a process of subdivision on an abstracted doric column.

In how many different ways can you sort a collection? From how many different perspectives can you view an assembly of things?

A collection is a group of things related to each other in some identifiable way. The relationship might be through a topic, a place, a person, a type of object etc. Even though there is a specific motivation for gathering on the first place, what makes a collection dynamic is that one can reorganize the same data in alternative ways and make different sense of it.

The motivation about this project came from the ‘Pink project’ by Portia Munson. The pink project comprised a series still life installations created out of the artist’s collection of discarded objects, all of which are varying shades of pink and are objects used by females. The discarded items assume new meaning when seen out of the perspective of the common color and the connotation with gender as well as through the way they are organized in space.

Portia Munson, Pink Project, 1994

The data set I am using for the project, comes from a collection of images of butterflies provided by an iPhone app made by Hunter Research and Technology. The collection is composed by 240 images. Each butterfly is shown in plan on a white background; the images I extracted are 260×340 pixels in size. The only data accompanying the images is the name of each butterfly. The sizes do not reflect the real life size of the butterflies; they were probably all brought to the same scale before being used in the iPhone app. I later found that the original images are coming from the following website:   http://butterflies.aa6g.org/ where also a description of how the butterflies had been photographed can be found.

1. perimeter of the butterfly outline 
2. area of the overall shape 
3. number of detected boundaries on the surface of the wings
4. the image’s average value 
5. the image variance 
6. color histogram

From the results I got for the images I noticed that the values do not vary significantly. This, I believe, is also reflected on the slideshow: in most of the cases the reason for transitioning from one butterfly to the other is not observable. My impression is that given the collection under consideration (same family of things with similar characteristics) the linear ordering might not make that much sense in terms of results.

Also, it might have been wrong on the first place to compute the image statistics without taking into consideration that a significant amount of the image pixels were those of the background. So I run the histogram analysis taking a different approach:

(1) I considered all three color channels

(2) I masked the image in order to compute a histogram only on the butterfly shape

(3) I computed the similarity among all pairs of butterflies and got the corresponding sorts; more specifically given the histogram vector for each image I computed the distance matrix between all pairs of butterflies.

The results seemed more reasonable. Thus, taking the three points mentioned below under consideration, I repeated the image processing and extracted the following image descriptors:

1. perimeter and area – considered together in one descriptor
2. color histogram
3. edge histogram – horizontal, vertical and edges on the two diagonal were considered for the computation of the edge histogram
4. average value – the average value was computed separately for each channel
5. standard deviation – the deviation was computed separately for the columns and rows of the image matrix for each of the three color channels
6. color histogram – 10 bin color histogram for each r g b channel

The following images are from my first attempt to visualize the data. I thought that it might be interesting to observe what kind of relations may occur among two different sorts. In the applet below I represented the sequence of a given sort e.g. by variation with a sequence of dots where each dot represents a butterfly in the sequence. By following the movement of a curve in time representing another sort e.g. by number of blobs I can compare how similar or not is the distribution of the butterflies with regard to the first sort.

I consider that this attempt does not work in communicating the objective of the visualization. I redesigned it following the rules below:

Project 1:

http://www.generatorx.no/category/generative-art/page/2/

Wow! Okay so this project is essentially the generation of physical objects from sound analysis. The idea, Cylinder, by Andy Huntington and Drew Allan is truly impressive because it takes information that can’t be seen, and generates these models as art. It’s ridiculous in how well these models turned out. Huntington and Allan go through and essentially use sound waveforms (usually a 1D graph) and spectral map (2D) to create this stunning 3D object. If I had to change anything, I would definitely use color on the objects to represent something. I know white is likely more practical, but color, to me, is always a good thing and one of those extra steps that in a lot of cases, I find definitely worth it.

Project 2

http://www.generatorx.no/20060108/math-lessons-for-right-brain-people/

I found this post on www.generatorx.no and essentially it deals with “Math lessons for   Right Brain people”. As a Math x Design BSA major, I tend to deal a lot with implementing Math in my Design. And occasionally, Design in my math. What people don’t always get is that Mathematics has a very open design. If you are proving theorems, proving continuity, or working with n-dimensional spaces, you’re implementing a certain design in the methodology and order of your work. This post is about a guy named Paul Falstad who created a view applets which generated art from mathematical theories and issues. Through his applet, Falstad is at least attempting to cross the bridge from Math into Design & Art. While not completely successful in my eyes, due to it’s lack of interactive meaning for the user, this work does symbolize one thing, as the article points out: “there is hope” for any student who seeking a creative understanding of a historically-non-creative field.

Project 3:

https://www.flickr.com/photos/stringdna/6867656741/in/photostream

While I don’t know exactly what this represents, it seams that Ben Guerrette’s image was primarily generated using an ActionScript project with some minor adjustments in Photoshop. This image is just gorgeous to me. To represent these colors, which may have been drawn from photos that Ben had, or photos stored on Flickr, as a generative art piece is just awesome. I’m a man of color. I’m a man of geometry. And honestly, this art piece is just breathtaking to me. Sure, it’s generated or contrived from data, but I find it unique in an of itself, which to me, deems this work as successful. If I had to change anything about it, well….I wouldn’t. The background lighting, the Circle diameters, the color gradients, the outlining of shapes, all of it just works. This type of generative is compelling to me because it allows the user or viewer to explore their own thoughts while their eyes traverse the numerous circles. Well Done.

Action Painting Redux

by Jeremy Rotsztain
Technology: openFrameworks
[vimeo=https://vimeo.com/26299391]

1. Hair Particle Drawing

In this rather weird project, the artist developed software that takes an image and redraws it… using curly hair patterns!
The original photo presented in the website is labeled “Bush”, and has George Bush’s image made up of hair. I appreciate the satire and the use of double meaning for the word bush.

The two rightmost photos show two stages of the process of generating such a photo. It is animated on the screen, you are more than welcome to check it out yourself and upload your own image. In honor of Valentine’s day, I uploaded a heart and thought it would be funny to see it made out of curly hair.

What I like about this project is that it combines two worlds – a static image of a person, or any other object the user wishes and a piece of curly human hair, and uses the hair to re-create the image. The new image is no longer either of the originals – it’s not a hair, and it’s not the image. It’s quite disgusting, if you ask me. I also like the fact you can load your own image and experiment with it.

The link can be found here. Enjoy!

2. SUBSTRATE

The following online gallery has lots of stuff that the artist created using processing. One that specifically attracted my attention was a simple algorithm that creates a city-like form, as can be seen in the following images:

I really like the stonewash effect the artist added to it to make it look cooler. I also appreciate the fact that an algorithm can create a rather random image, that will look like a city map to us. Makes me think about the elements that compose a map, and as clearly this is not a real map depicting a real place, what in our mental model is “a map”, and how does this generated image fulfill that.

You can play with it, it is addictive! Go here.

Also, from the same artist, I had to add another one.

3. Hyperexperience

For my third “looking outwards” item, i chose a non-digital form of randomly generated art. Not only it is non-digital, it is also crowd sourced. The artist created an installation of a wall which you can tear up and stick things of you own. I am always excited to see these crowd-sourced project come together, as they seem to have no one managing them or directing them, but somehow they maintain a common language and refrain from becoming just a random mess. Introducing human behavior as your random component, brings some beautiful results!

Check it out over here !

VALSE AUTOMATIQUE PROJECT| MADE

Stephan Thiel was in charge of most of this project they describe as “design performace.” The final piece is 5 individual carved wax cylindrical sculptures. Each sculpture individually crafted as a visualization of the composed music.

nº 502 – 519
By: Pe Lang

Pe Lang creates incredible work. I discovered Pe in the new issue of Neural. His work is simple visual and he has commented saying that the simplicity of the work aesthetically is integral to getting his messages across. He creates machines that are visually simple yet executing complex manipulations of parts. This work is both a visualization of sound and generative motions.

Joshua Davis:

Josh has been one of the earliest designers in the world of generative design. His work has been featured in every type of setting possible. His roots are in the west coast street art and his work is obviously representative of that but yet his work is obtainable enough that companies like IBM’s Watson Computer and now Prius Hyrids have hired him to create generative designs.

Wim Delvoye – Cloaca no 5

This work is a piece of shit. No, rather, it creates a piece of shit. Cloaca No 5 is a mechanized digestive system. It takes food items into one end and processes them through enzymes and acids meant to replicate the human digestive process, resulting in a simulated hunk of fecal matter. The shit is then packaged and sold on the art market as a rare edition.

Tissue Culture and Art Project – Victimless leather

If a piece of leather is created without hurting or killing an animal, is it still ethical? Is meat that is produced from tissue engineering vegetarian? The Tissue Culture and Art project sets out to raise these questions by creating simulations of meat and leather through tissue culture engineering. By deploying tissue cultures on organic scaffolding, these artists were able to create simulations of products that otherwise would require the maiming or killing of animals.

Alvin Lucier – I am Sitting in a Room

In his 1969 recording I am Sitting in a Room, Alvin Lucier deploys an algorithmic method to generate an audio recording based on the environment it is recorded in. Lucier’s process is as follows: He records himself speaking a simple sentence (“I am sitting in a room…”). He then plays back that recording, and records the sound of playback in the very same room. He repeats this processes through several iterations, each of which amplifies the resonant frequencies of the room as they are resonated by the sound of Lucier’s voice. By the 6th or 7th iteration, the sound of Lucier’s voice is almost completely diminished, and all that is audible is a cascading raga-like harmonic, pulsing to the rhythm of his speech.

While on the look out for generative or simulation-driven pieces, I found too many that I thought were interesting, so this post is rather full.

Generative Color Paintings

Two different approaches to generating digital paintings exploring color and pattern.

Watercolor Sediment

by Justin Livi

I think the results speak for themselves here. The watercolor effect is well done and the color variations are visually appealing. I don’t have much else to say, other than I’d love to have a wall-size print of one of these.

RGB Petri

by Jeremy Awon

[vimeo https://vimeo.com/699396]

First, this is simply fascinating to watch because the image grows outward from a single seed point. Second, the use of pixels as the fundamental “cell” leads to an interesting stylistic change at different scales. At the “normal” view, the result is very organic, almost a watercolor effect like in Justin Livi’s project. But when zoomed in, the images have a wonderful “retro videogame” feel (see examples here). I do wish there was a way to set the starting color in the applet or have multiple seed points; I think the red is too dominant after several images.

Both projects discovered via CAN.

MIT Media Lab Identity

by The Green Eyl and E Roon Kang

[vimeo https://vimeo.com/20488585]

I think this is a brilliant use of generative techniques. The designs are consistent and clearly fit together, yet each person associated with the lab is able to design their own personal version of the logo. I also like how the generation technique easily allows for animation of the logo in video. But what I find most interesting is how the the sparse use of geometry is able to integrate many different color schemes; while shape is most important in a logo, I’ve always assumed that color was almost as important. Appropriate that it was developed in Processing.

Word and Title Generation

Two projects working with generative text. The first is interesting on a conceptual level; the second is included because the results are so entertaining.

Random Word Machine

The random words are interesting because most are pronounceable and could be real words, but they would be decidedly strange if they were real. The machine seems to prefer infrequent letter combinations, which works well to make unique words, but also leads to things like “erbagwric” or “gonvexismy”. Also interesting is that many words seem to contain actual words or pre/postfixes. I find that reading these smaller words determines the meaning of the random words to some degree; “gonvexismy” must relate to confusion somehow.

Video Game Name Generator

This is pretty silly, but impressive in that the titles are both believable and completely ridiculous. There’s not much else to say, so I’ll end with some examples:

• Legendary Grizzly Bear Fever
• Professional Kangaroo Hunt
• Disturbing Karaoke II
• Biblical Monster Truck Assassins

Nervous System

Nervous System is a design studio that uses generative art algorithms like DLA (Diffusion Limited Aggregation) to create 3D biological patterns for jewelry & decorative pieces. (link) Their code is also open source, if you want to use it. The best part for me is their idea of being able to customize the algorithm parameters via an iPad app and personalize the jewelry you would like to buy and still keep the prices reasonable & accessible.

Cylinder

Cylinder is a project that takes sound frequencies and renders them in a tangible and permanent manner as sculptures to represent a sample of time. (link). I think the output is beautiful and it is representative of the way sound could be visualized. The final sculpture is white and takes only one thing into account: I wonder if using color might result more informative yet prettier sculptures.

Written Images

Written Images is an experimental publishing concept where each copy is generated using algorithms so no two copies are the same, but all are within a tight enough possibility space to be regarded as the same publication. I like the concept of a generated book: it kind of reminds me of game books where you can navigate the book (pages) for your own unique ending. Although this project generates unique images for each page, I can imagine a system that created a unique (text) book for you in a generative way.

SCIGen

Finally: I don’t know if this counts as generative or art but it is amusing. SCIgen is a program that generates random Computer Science research papers, including graphs, figures, and citations. Some of the papers it generated were published, including one on Interactive Media Design. [link]