As·phyx·i·a is my experimental film created with Frederico Phillips and performed by Shiho Tanaka.
Conceptually, as·phyx·i·a is meant to be an eloquent choreography that stresses the desire to be expressive without bounds or any common constraints in commercial jobs.
I was responsible for co-directing the project and created an original sound track which guided the narrative. The track was created using various digital and analog techniques. Frederico Phillips created the visuals, from pre-visualization to final rendering. Samuel Blalark assisted the project in several areas with key feedback and on location support such as filming for behind the scenes.
More info: asphyxia-project.com
There was a need to do something exploratory and experimental that satisfied our own curiosity and helped us answer some ‘what ifs’ and ideas we had previously contemplated.
The capabilities of Kinect Xbox One latest sensor to capture at a higher resolution then its predecessor caught our attention so we decided to test both versions against each-other.
We found out the new sensor was not only able to capture more detail but if we combined two sensors to capture both back and front we could then merge them together in a 3D application.
By allowing the physical back into the digital we had hoped the result would translate itself into a piece that people could relate to, even though it was fully computer generated it carried human emotion.
An environment was also built to integrate the character and allow for a more visually photo-real representation of the scenes.
We motion captured Shiho Tanaka’s dance using Kinect Xbox One, a sensor that is sold by Microsoft for the purpose of gesture based gaming.
It took us about 5/6 weeks to complete the bulk of the work and the longest part was actually waiting for the rendered sequences (over 5000 frames) as we only had 5 computers rendering the whole project (producing several image sequences).
There were many challenges as most of the tasks were new to us, such as finding suitable ways to combine and clean the data captured, producing a sound track that could help drive the visuals and coming up with a look that connected well with the concept or selecting aprox. 3 minutes of performance down from the original 30 minutes we captured.