The system I chose is the the “Facet” algorithm located in the Photoshop filter gallery. What I did is that I opened a picture in Photoshop, then applied the “Facet” filter onto the picture over and over so that the color and the composition of the pixels are transformed into something very different from the original.
It’s interesting to see how the result of this feedback system does not look like what I predicted at all. Please see the video documentation below to find out how the picture has evolved. The filter was applied for 1000 times.
This is a video that shows how Google Translation can distort the meanings of a poem by recursively feeding the translation results back to itself and translating to random languages 60 times. The original poem is in Chinese, and very well translated into English. The title is Facing the Sea with Spring Blossoms by HaiZi, a Chinese poet.
From tomorrow on, I will be a happy man.
Grooming, chopping and traveling all over the world.
From tomorrow on, I will care foodstuff and vegetable.
Living in a house towards the sea, with spring blossoms.
From tomorrow on, write to each of my dear ones.
Telling them of my happiness.
What the lightening of happiness has told me.
I will spread it to each of them.
Give a warm name for every river and every mountain.
Strangers, I will also wish you happy.
May you have a brilliant future!
May you lovers eventually become spouses!
May you enjoy happiness in this earthly world!
I only wish to face the sea, with spring blossoms.
After 33 random translations:
ur
You know you went to the world,
Cooking vegetables and vegetables
I am a source of water and water
In the morning, family relationships
Stories, history
It’s a good vaccine
Everyone says,
He was always his name
I do not believe
if you want to
Hence
the poor
Chicken in the ocean
After 60 random translations:
road
You know, I go in the world
All full of fruit
Water and Water Resources
It is very close to me in the morning
History, history
It is the perfect solution.
a total of
It is always the name of the
I always thought,
If you want to
yes
bad
Poland
This is the opening sequence of Alfred Hitchcock’s film, Rear Window. The original digitized (via YouTube) film clip was played on my computer screen using MPEG Streamclip and simultaneously screen recorded using QuickTime. The newly recorded version was then played via MPEG Streamclip and again screen recorded using QuickTime, and so on.
As the signal decayed, the video slowed, stuttered, became higher in contrast, and eventually darker and darker, resulting in the interesting illusion of the scene taking place later in the day with each subsequent rendering. After 18 iterations, the film became almost completely black, with only the appearance of what seem to be small fiery flares visible in the darkness.
The grid below is a composite of all 18 passes simultaneously – best viewed in full screen!
The above video is the result of a beautifying filter from a photo retouching app (Meitu) being applied to the same picture (of myself) about 120 times. The filter itself was the one the app called ‘natural’, set to be as subtle as possible. (A slider in the app alters how drastic the changes are.)
To follow the themes presented in the in-class examples, this project was meant to demonstrate certain fundamental artifacts and assumptions in the process itself – in this case, what does the app consider beautiful? We see in the final result of the process those artifacts extracted and brought to their extremes.
The music on the video doesn’t have anything to do with feedback loops, I just thought having it in the background was funny.
The system I selected was the information database, Wikipedia. I started with a simple search term, Signal, and followed what I thought were interesting links from page to page. Many times I have gone down the wormhole that is Wikipedia links, and ended up nowhere close to the topic I started off on.
This particular trip took me from internet history, to U.S governance, back to computer science, to Carnegie and Mellon, to golf, to horse shows, to obsessive compulsive disorders, to polio, to history, to math, to data, to film, to mythology, to national disasters, and finally currency.
The system I chose to work with was Youtube. I clicked on the #1 video in the trending section, and it was “Taylor Swift – …Ready For It? (Audio).” I then clicked the 5th video in the up next sidebar. I repeated the process 25 times, but prevented videos that I had chosen earlier from showing up. The 25th video was “Teen Titans Go Transforms Baby Raven Starfire Growing Up Surprise Egg and Toy Collector SETC” which had no relation to Taylor Swift, or even music for that matter. Through this process, I destroyed any evidence of the original video through feedback.
The system I chose was an online thesaurus- Thesaurus.com. The original input word was ‘candy’. This word was chosen randomly. Once this word was fed into the system (the thesaurus), synonyms were generated. The first word from this list was chosen and fed back into the system. One rule that I applied to prevent an infinite loop was to exclude any word that I had already chosen earlier. The end result was a word that had no relation with the original word whatsoever. Thus, destroying the original word and its meaning through feedback.
I used two phones to distort an image I took a couple years ago. Using one phone, I took a picture of the image displayed on the second phone. Then, I used AirDrop to quickly transfer the file to display the new image on the second phone and repeated the process until the image became mostly one color after 50 iterations.
Assignment 4: Due October 16 – create a system that does processing in the frequency domain using [pfft~]. Submit your code and documentation of your system/instrument in use.
Transform an audio signal by convolving it with four different Impulse Response recordings. You should make at least two of your IR recordings by using portable audio recorders to record the reverberations of a “pop” in two different acoustic spaces. Try to find unique acoustic spaces that will create interesting reverberations. The other two IR recordings can be more experimental. For example, one can got interesting results by treating musical sounds as if they were IR recordings.
To deliver your work present:
• The original signal
• The original signal convolved through 4 different IR recordings
• The 4 IR recordings, and a brief description of how they were produced
Convolution of images is acceptable as well if you’re interested in doing a visual version of this project.
