Partner: Keer Zhou
P.L.S. is an AR installation exploring questions of identity: who is the individual in an era of digital globalization? Through a mixture of interactive audio and visual elements, the audience will explore a mesh of floating cubes within a doppler radar backdrop. Each face of the cube is composed of parts of a face, alluding to the packaged social profile sent out into the ether. The audience is invited to approach each cube to unpack its story. P.L.S. stands for Packing, Loading, Shipping.
For our exquisite corpse, we decided to make a cube composite of parts of our two faces. My partner Keer made the following cube in Photoshop:
Our original idea was too ambitious and not well-suited to AR (in summary: to cube is part of an animation sequence wherein the faces start off as tiles of an Instagram profile; they fold into a cube which goes onto a conveyor belt, which then sends it off into the ether as if being shipped out. We went through a number of iterations, which you can see on this blog post).
Once we explored After Effects, we learned that our idea was too ambitious to execute in After Effects. Within the software, we instead decided to simplify the animation to just have the cube fold, float up, then twirl. From here, we would import it into Adobe Aero for its AR component.
Even creating the folding animation in After Effects was tricky to do well. It was very time consuming to get all six images which made up the sides to fold the correct way and then align as a cube. Here is the cube at the point at the point where I stopped fiddling with the fold:
Screen_Recording_2021-11-09_at_00.37.58.mov
It was really difficult and finicky to try to get all the sides to align! And there wasn't a clear way to do this with the software. Only later did I find out that there are assistive features in the software which can help with this, but even then they didn't work as smoothly with individual Photoshopped images as they did with a shape drawn in After Effects. (Could we have drawn an image and then made the shape fill be individual images? This is something I thought to try if I were to do this project again).
I then added a feature where the cube rose a bit because we wanted it to float above the surface from which it emerged (ideally a phone), with the folding animation changed a bit:
Screen Recording 2021-11-16 at 00.10.56.mov
I did this by changing the position of each photo by -416 units. This wasn't the most efficient way of doing it, and I only later discovered that if we had made each image be the fill for a shape, instead of an individual photo, I might have been able to combine each one more easily.
Finally, I added a twirl. I combined the layers into a single shape in After Effects. However, it did not allow me to turn this shape into a 3D object. This meant that I couldn't whirl it, and in AR it didn't appear 3D. This was a shame, but I couldn't figure out why this was the case. I twirled it on its center axis instead.
Screen Recording 2021-11-16 at 00.23.46.mov
However, once I shared the folder with this work with my partner, I learned that she was not able to open it in order to work with it in Aero.
In all, it was quite difficult working in After Effects. There didn't seem to be universal ways of doing things, and I spent a lot of time trying to navigate and make sense of the layout and functions in the software. Even after hours of working in the software to try to make a simple cube fold, float and spin, I don't have a solid foundational grasp of how the software is structured. The fact that a simple folder share didn't work when I shared it with my partner is a disappointment.
My partner eventually imported the cube into Adobe Aero and we discovered that it was much easier to enact the folding animation there. However, the cube only appeared to be 3D; when the user would try to go behind it in AR, it was clearly 2D.