Visualising motion data
This project was developed for the 2015 Choreographic Coding Lab that took place in April 2015 in Melbourne at the Deakin University Motion Lab. The motion data was performed by Niharika Senapati using the Countertechnique dance method developed by Anouk van Dijk.
My goal for the 4 days of the lab was to look at visualising motion data in non-traditional ways and to create new standalone artworks interrelated but independent to the performance. I hoped to provide some different perspectives in looking at motion data for choreographic, teaching and reflective purposes.
Motion Data as Paintings
I started tackling the motion tracking data by isolating specific body movements, instead of trying to piece together the whole human form as a way of extracting an abstract view of the data. These are the initial experiments with drawing out the data points as 2D representations using a birds-eye perspective. I was really impressed with how smooth the lines came out, most likely to the credit of the Countertechnique methods.
Next, I played around with exploring other properties of motion data, one of which being velocity. With this sketch, I wanted to use the body like paint brush, with increase in velocity corresponding to an increase in brush strokes.
This last sketch visualises what it would look like if there were lines connecting opposing limbs. Connecting lines between opposing extremities; right wrist and left ankle, left wrist and right ankle. The results were the most abstract out of the sketches, but exposed a perspective on the relationships between parts of the body that would otherwise have been missed.
Motion Data animated
This series of animations show the sketches playing out the dance movements.
For the last sketch, the end result still feels like a dance, but resembles more closely birds dancing in flight.
I created a tool to load in all the motion data files and be able to watch them play out using the different visualisation techniques and at different speeds. Play with it here.
Thanks to Scott deLahunta and Florian Jenett from Motionbank for running the Lab. Also special thanks and a mention to Phil Boltt for making the data digestible for a non-3d-savvy coder. Please check out his project using the Countertechnique data.