touch
December 18th, 2008 admin
Result of a quick exercise where we were challenged to create a visual representation of ‘touch’ [time restricted in-class project (25 minutes) ].

Posted in home, visualizing the five senses | No Comments »
December 18th, 2008 admin
Result of a quick exercise where we were challenged to create a visual representation of ‘touch’ [time restricted in-class project (25 minutes) ].

Posted in home, visualizing the five senses | No Comments »
December 17th, 2008 admin
This java application was developed to enhance the acoustic research myself and my collaborator Alberto Tafoya have been pursuing this semester. In short, we are trying to understand and control acoustic interference patterns.
This post will focus on the current state of the research application that was developed in Java and SuperCollider (real time audio synthesis environment).
The majority of our research has been conducted in less then ideal acoustic conditions, in spaces lacking symmetry among countless other contributors of error (windows, unstable temperature, reflective walls). These conditions make the task of accurately predicting the locations of acoustic phase cancellation significantly difficult. To account for these sources of error, the decision was made to programatically map the formation of the interference patterns through use of an 8 channel microphone array. This would allow us to modify the phase of each output channel and monitor the changing interference pattern through the amplitude values recorded by our microphone array. This form of feedback system allows us to work with the inherent imperfections that are present within our working medium.
The application serves two primary purposes. First it allows for the execution of lengthy automated tests where the resulting data is collected in stored digitally. An example of this type of automated test would be the mapping of the 360^4 combinations of potential phase shift combinations given a maximum resolution of 1 degree.
Once this data has been collected, this same application allows for the immediate visualization of the acquired data. The interface allows one to overlay the microphone amplitude values overtop of a mathematically correct prediction of the interference patterns that would occur within an anechoic space.
The image set below describes the various states of the application in its present form. The main display area allows one to see both the mathematical prediction of areas of phase cancellation as well as the microphone amplitudes gathered during a given experiment. The fifth slider is attached to the experimental data and forces the phase of the 4 channels to match the experimental conditions at which the visible instance of data was gathered.
[ please contact for source code and experimental data ]
Posted in home, visualizing the five senses | No Comments »
November 17th, 2008 admin
Wine has an associated descriptive vocabulary that is unparalleled against most other comestible subject matter. Structured, oak, plummy, nougat, toast, mulberries, acid and yeast are but a few examples of the diverse range of words that are actively used to describe the complex flavors of wine. When presented with a large set of wine tasting notes that throughly described over 5000 wines, I was immediately drawn to the associated prices when reading through the tasting notes of a particular selection. Due to the financial realities that I have come to terms with, consuming a bottle of wine beyond the $20 dollar threshold is a rare occasion. Given this circumstance, I was personally drawn the descriptions of the wines that ranged in the hundreds-of-dollars out of sheer novelty and curiosity. This curiosity lead me to then question weather there was a hierarchy of vocabularies within the wine descriptions themselves. While reviewing the wine selections within my budget, was I not being exposed to the full gamut of descriptive words? Is there a premium on certain adjectives when describing such a well understood and valued commodity?
This visualization allows a user to first select a target price of wine. When a price has been selected, the entire descriptive dictionary used in the tasting notes of all 5000 wines is queried to return only those words that have described a wine at the user specified target price (+/- 5%). One can then view the distribution of these now relevant words across the entire price range of wines. For example, if a user defines a target price of $133, only the words used to describe wines at that price will be pulled from the complete dictionary. With a quick glance of the distributions one can infer an understanding of how this subset of words is used across the complete price range of wines. Notable results from this specific query are the words ‘Feminine’ which only appears at the $127 price point. Conversely the word ‘with’ is present across the entire price range.
Additional features have been built into the visualization tool to enhance the exploration of this data-set. Beyond the prominent mouse controlled red scroll bar along the top right of the window that allows for a target price selection, one can scroll vertically through the data-points to reveal the dollar value of the point that one is mousing over. Further more, a single click of the mouse anywhere within the visualization window will pull an additional 55 entries from the relevant subset of words (when applicable). Due to the broad spectrum of wine prices, gaps do exist where a target price will return no words. In this case, one simply has to continue navigating across the scroll bar until a target price is found that returns a subset of descriptive words.
The images below illustrate a few sample cases of the visualization.
An executable java application in *.jar format can be downloaded here.
Posted in home, visualizing the five senses | No Comments »
October 21st, 2008 admin
Posted in home, visualizing the five senses | No Comments »
October 7th, 2008 admin
///////////////////////////////////////////////////////// PART I
3D representation.
concentration of odor source is constant across the Y axis.
percentage of time in sector represented by Z axis.
similar to:

![]()



///////////////////////////////////////////////////////// PART II
Posted in home, visualizing the five senses | No Comments »
September 30th, 2008 admin
As described in the proposal, this assignment allowed me to further explore the imperfections within the vinyl medium. With the week we had to prepare this visualization, I was able to extract the desired information through the analysis of the captured video feed. The data acquisition relied upon the blob-detection library (java). Modification of the library was necessary in order to cater to the specific needs of this visualization. Most significantly it was necessary to track a single ‘blob’ in order to correlate a fixed location on the vinyl cartridge (from the video data) in order to monitor its movement on the X and Y axis. Limitations within the blob-detection library prevented such localized tracking before the modifications were made.
Once the data was extracted, the average vertical and horizontal translations made by the cartridge were calculated. These average values were then subtracted from the cartridge position at each frame, leaving only those exceptional or abnormal movements of the cartridge to be highlighted. At this point, there are no empirical values representing the movement of the cartridge, however each un-expected movement is represented as a percentage of the total distance traveled by the cartridge for a given data set. The red crosshair marks the location of the cartridge every 50 frames allowing for a ’snap shot’ of the given deviation. The resulting visualization can be found here.
The image set below first illustrates the data acquisition process. Specifically one can observe the difference in the data quality between the modified, and unmodified blob-detection library. The remaining images are of the visualization in its most recent manifestation. Please note this project is still in progress.
Posted in home, visualizing the five senses | No Comments »
September 23rd, 2008 admin
providing a vivid perspective into the subtle nuances of our everyday life, i am looking forward to interpreting the output of a low powered digital microscope. after spending much time examining the cracks and crevices of my laptop, the movement of loud speakers, clothing, tables, cellular phones, monitors, vinyl records and fluorescent lighting, I was able to comit towards a target for this next visualization exercise. recently i have been fortunate enough to spend significant amounts of time listening to vinyl records. personally I find the process of handling physical artifacts of music, manipulating the player with bare hands, and the visual spectacle of following the needle slowly traverse the radius of the record to be an integral part of the listening experience. of the most striking elements of this process are the clear imperfections that are occasionally heard through the speakers as ‘clicks’ and ‘pops’ due to dust and expected wear and tear on the audio equipment. this visualization will however focus on the physical imperfections manifested through the vertical and lateral movements of the cartridge as it slowly scans the the vinyl surface of the record. this will allow me to explore and display both the physical tracks laid out on the record and the tool used to transcribe the topological information. again, there will be a specific focus on outlining the vertical and lateral movements of the cartridge as it travels through the less then perfect circular tracks that comprise the auditory information on the vinyl disks.
this visualization will highlight both the imperfections within the medium and the physical compensations made by the interpreting equipment. given a fixed viewing position, the visualization will represent the movement of the components in comprehensible units allowing for the subtle nature of this space to be better understood. from an esthetic perspective i intend to generate a highly technical overlay that will tediously analyze the previously discussed nuances of the medium.
the image set below provides a few sample perspectives into the space I intend to describe through this assignment. although the camera was not secured by a tripod while these images were captured, one can get a feel for the movement of the cartridge as it traverses the imperfect path of the record used in this shoot.
Posted in home, visualizing the five senses | No Comments »
September 18th, 2008 admin
Building upon the extremely minimal spectrum analysis from the previous week, I decided to expand the application through the addition some GUI elements allowing for a more precise understanding of the analysis. Specifically, the addition of two slide controllers that permit the user to define two key components within the FFT analysis. The controls in question include the ability to distinguish the minimum octave width (measured in Hz) and the number of divisions that each octave should be subdivided. The application is initialized to create 30 frequency bands (minimum octave width 22Hz, with each octave divided into 3 bands). By increasing the minimum octave width one is essentially reducing the resolution of the analysis. At a maximum octave width of 6000 hz, the number of frequency bands is reduced to 5. At this point, only the left-most of the frequency bands is stimulated as the majority audible range is found within this now augmented frequency band. Conversly one can increase the resolution of the analysis by increasing the amount of divisions per octave. The image set below provides a general overview of the various manifestations this analysis can represent.
Posted in visualizing the five senses | No Comments »
September 9th, 2008 admin
a fast fourier transfrom is used to separate the audio signal into 30 distinct frequency bands. the amplitude of each frequency band is calculated and directly mapped to the RGB color values of its associated bar within the visualization. as the amplitude of a given frequency band increases its associated bar color is proportionally modified. highly active frequency bands will appear white in color, while less active frequencies will remain black or gray. as the amplitude of a given frequency band changes, color modifications are layered atop the existing representation with an opacity of 5%. this generates a gradual change in regards to the coloring of the individual bars. the color pallete in the sample images below are not final, and are being used simply for testing in the early stages of this project. the represented spectrum ranges from 0 Hz on the left most extreme of the visualization to 22000 Hz on the far right.
Posted in visualizing the five senses | No Comments »