Wednesday, September 1, 2010

Dynamic Clothing Prints

As part of a research project earlier this year, a colleague and I developed an idea about the use of dynamic garment prints which we presented to a scientific conference concerned with a special kind of computer data structure (called the Voronoi diagram). I often don't think to connect my fashion/sewing blog to some of the more relevant research projects with which I'm involved. In the future, I'll try to remember to do this more, as I think some of these projects may be of interest to people interested in fashion design.

Our original idea was to use this particular kind of data structure as a support for software that could be used to combine the stages of printing, layout, garment construction and garment wearing. Using modern animation techniques such as dynamic cloth simulations, it is becoming easier to link the garment construction and the overall look of the garment under motion, and existing software packages allow one to specify fabric and layout and infer garment construction. However, incorporating the printing stage is still very difficult to do, even though, with digital fabric printers, it is now possible to custom design prints to create the final garment appearance. Right now, the process of getting from the printing to the appearance of the final garment is very much by trial and error. It would be great to have access to software that would allow one to "slide the print around", or make changes to the print and see instantly the effect on the final garment.

There are a number of technical problems that have to be overcome to be able to do this. Although not the only solution, and perhaps not even the best, the Voronoi data structure offers some flexibility in being able to do this, which is why we investigated this for our research.

However, once we got onto the idea of being able to "slide around and update" a fabric print within the garment layout, and see instantly the result on the final garment, we realized that a similar procedure might be used to generate "dynamic patterns" that change in real time on the garment's surface. This rather cool idea, requires new technology to actually implement - we are not (yet) at an era where fabrics are a kind of flexible computer screen, which is what one would need, ideally, to achieve such an effect. However, with a little imagination and technical know-how, it is possible to conceive of a number of lesser effects that would still provide interesting garment experiences.

(a) One way to do this would be to equip a garment with an array of LEDs (light emitting diodes). LEDs are cheap (less than 2 cents a diode) and there are now a number of technologies that allow them to be integrated into clothes. Although controlling a large array of lights is still quite demanding, controlling a smaller number (a dozen or so) is relatively straight forward, even for the amateur. Over time, controlling arrays of LEDs should become easier. However, this is perhaps not the best solution.

(b) The use of chromothermic inks provides another means to dynamically change fabric prints - however, the principle is fairly constraining. Essentially, different colored inks can be used that become transparent at different temperatures - for example, red at 15 degrees celsius, blue at 16 degrees, green at 17 degrees and so on. By heating the garment with a low voltage electric circuit (or simply as the ambient temperature changes), the garment's printed pattern will change its appearance. A number of new garments presently emerging from experimental stages are using this principle. However, for dynamic garment prints, the applicability is still limited.

(c) An alternative approach, and one closer to our interest, is to use augmented reality. The term refers to the idea of viewing a camera image which shows the "world" as it is, but overlaid on top are virtual objects, which, in sophisticated AR applications, appear to be part of real world objects. Traditionally, the use of AR involves wearing specially designed glasses or goggles that incorporate the cameras, but recently the iPhone and similar devices has been used to provide a similar capacity, not "glued" to our eyes in the same way. That is, when you look "through" the iPhone, you see the world "augmented" by the virtual elements.

(d) A fourth possibility, related to the AR approach, is to project images onto the garment. This could make sense, for example, for a play or dance show that one watches from a distance - the effect is limited to a particular viewing direction. This has been done on several occasions by avant-garde artists, so it is not particularly new, although before the arrival of Augmented Reality software modules in the public domain (of which there are now several), this effect was very challenging to achieve!

To explore what this might look like (either option (c) or option (d)), we simulated the use of a dynamically "corrected" print using a virtual development environment (Poser 7). In the video segment below, you will see a virtual woman wearing a blouse with a static print (the usual!), lean forward and then back, followed by a sequence showing the same movement using a dynamic blouse pattern. The idea is to note that the dynamic blouse pattern leads us to view the body differently than does the static pattern. There is a bit of a "moiré" effect, due to the single perspective and the "perfection" of the dynamic pattern - in a real implementation, one would not expect to see this effect, as there will be many small imperfections in the way the pattern will look, but overall the different perception should still be maintained if the print changes over time.

Of course, there are many other ways to make the print vary over time - in our work, we were particularly interest in body image and body perception.


  1. The system in (b) could be applied to a tarp to be placed over a vehicle or other item that could be dynamically camouflaged. This could have some great defence potential - and lots of $$ for R&D. (Ultimately, in uniforms, this would be great. I just figure a tarp would be an easier, highly flexible easy to demonstrate place to start.)

    Just a suggestion. :o)

  2. You're right about the potential for defense applications - not the first time I've encountered this with regards to my research. But not my cup of tea. I shall leave this up to others to develop :)