Visualizing Motion
Real-time Motion Tracker via Fologram
Course: SCI 6338 Introduction to Computational Design
Instructor: Jose Luis
Collaborator: Ana Loayza, Haochen Zhang
Date: Sep.-Oct., 2020
This project explores new ways of integrated form-making and cosmetics through the production of columns by means of motion tracking.
Two devices are tracked to obtain ‘proto-spirals’ that are later manipulated by scaling, arrays, and object placement to shape new columns.
KEYWORDS:
Form-making, Motion Tracking, Fologram, Column
OBJECTIVES
1. Record a few motion gestures using the provided template files and documentation.
2. Obtain a CSV file with the recording.
3. Write a Grasshopper definition to read the CSV file and turn it into the source geometry or use the completed one provided with the files.
4. Extend that file with your own implementations to generate your desired visualization.
Hacker level:
5. Use multiple devices for the recording, and plot them on the same visualization!
6. Real-time performance: connect the recording scripts to your visualization ones, to be able to generate a real-time performative piece.
Inspiration
Motion-based:
Form-making and Cosmetics
Motion Tracking Planes
Precedents: Michael Hansmeyer’s cardboard columns (left) and ‘Concrete Choreography’ Concrete 3D Printed columns (right)
Spinning: We were inspired by the Figure Skating sport, trying to record the position of hands relative to axis of human body.
Methodology
Fologram Motion Tracking
Click here to learn more about Fologram
Experiment
Motion-based Columns: Skeletons
Experiment
Motion-based Columns: Results
Experiment
Column Making Strategies
1. Loft (type 1)
Lines (AB) are generated from planes of two devices (A and B). Loft is applied to the list of curves
2. Loft + section (type 2)
Lines AB are used to obtain a midpoint and +- normal. New lines (A - Midpoint) and (B - Midpoint) complete a new section. Loft is applied.
3. Polar array (type 3)
GH C# component is used to generate new iterations from Loft (type 1). Different number of elements and angles are used.
4. Other variations (type 4)
With object placement along new interpolated planes. Cones and boxes are used to generate columns from the array of elements.
Experiment
Column Detail Mosaics
Experiment
Manual Configuration
Column 1,2,3
Array of original loft with different number of elements and rotation angles.
Column 4,5,6
Columns 4 to 6 are modeled with the object placement script. Sequences of objects A and B become inputs for the array component.
As opposed to the columns 5 and 6. The column at the left side is made with rebuilt lines A and B with less control points. Also, boxes have bigger sizes and lower number array steps.
Experiment
Real Time Motion
Experiment
Advanced Visualization
Color Change Dance
Column Fire Flame A
Column Fire Flame B
Discussions
1. Column making required several steps from purging csv data, scaling planes position and applying techniques to achieve levels of massing that allow the final volume to be read as a single object.
2. The original csv files would come with patterns of data that required to be purged.
3. The technique of adding a new section to the original lines AB allowed to achieve thickness in the original loft . However, it is effective until the new section generates visual conflict on its massing, a variable that depends in the csv data and varies from file to file.
4. The outcomes from arrays depends mostly on the rotation axis, which is required as an input for the GH C# array component. It is an effective way to create perceptible mass. However, the final massing as a single item – and not as a compound of meshes- is matter of further work.
5. Although structural performance is not a major criteria in this exploration, the team has detected some incongruencies in the final form of some columns such as folding conflicts, voids among elements that can be interpreted as a structural discontinuity and other kind of anomalies. Paradoxically, some of these anomalies imply at the same interesting cosmetic patterns integrated to the general massing.
6. Columns up to this point could be in beta stage.