Untitled-4-01.png

[m] Algorithimic Masks

Algorithimic Masks

 
 

Focusing on surface manipulation, discretization, and curve processing, this project aims to create architect’s masks using algorithm-generated geometry.

 
 
 

Project

SCI 6338 Introduction to Computational Design

Instructed by Panagiotis Michalatos

Tools

Rhinoceros, grasshopper, c#

Partner

Boxiang Yu

Keywords

computational design


STEP 1 ACQUISITION

What is a mask for architect. A mask is defined as a covering for the face and by looking at the artistry of Bjork’s performing masks and runway classics, and we were able to define that a mask for architect might suggest the prominent works of his or hers while retaining the basic anatomy of ones’ facial feature.

We were really inspired by an illustration that portraits iconic building geometry by each architect. By zooming in the the design style, we have decided to select Gerrit Rietveld, Louis Kahn, Daniel Libeskind, Frank Gehry, and Richard Meier’s works as our base lines. Each architect’s style has been simplified into one single line which will help deform and re-sculpt the facial construction.

Screen Shot 2019-01-08 at 11.41.33 AM.png

Image source: online

STEP 2 deformation / resculpting

In order to deform the face using the line information we have generated, we use C# code in grasshopper to capture the grid line of the face geometry. Then by placing the line near the face, we adjusted the code to read the distance from the line to the face grid, and then by transforming that information to the shift at the Z elevation, we were able to generate an interpolation surface, aka the newly sculpted face from the lines that’s representing the architects, from the points that’s affected by the line and the face surface simultaneously.

step 3 discretization

We have applied contouring, adaptive subdivision, offsetting, and object mapping to the discretization step. By offsetting along the u-axis, we were able to deform the geometry more abstractly. Furthermore, we have applied contouring to the geometry to change it into a surface that is more manageable. Using subdivision we were able to subdividing the surface by detecting the height difference between the surface grid. When the tolerance is below the allowable range, the surface will subdivide, so that we could learn how constructible the face could be. By offsetting the selected subdivided surfaces and applying a new geometry to the face, we had the final discretized mask.

Screen Shot 2019-01-08 at 11.44.21 AM.png
 

MASKS

 
 
Screen Shot 2019-01-08 at 11.46.48 AM.png
 
 
 

code