It all began because I needed pants. No, well, it all started because my old pants split. Actually, it all really started when I watched Waking Life and became obsessed with making rotoscope-animated short films of my own. Trouble is, that is painstaking back-breaking hard work of the kind one may consider undertaking if someone offers to pay your bills. I pay my own bills so that kind of work is not something I undertake lightly. Of course you don’t understand how tough a task can be until you undertake it so I made one weak attempt in the style of Richard Linklater’s film in the months after I first saw it. And that was that.
In the recent past I became aware of the smartphone app Prisma and I even made a few photos (illustrations?) with it. But I thought that the real value of Prisma–for filmmakers like me at least–would be in employing their technology to ‘animate’ live action footage. I even remember reading that they were working on a video version of their app but that version isn’t here yet (right?) and I wanted to add a little (nay a lot of) zing to the video we had made for Day 62 of the 365 Video Challenge we embarked upon at the beginning of 2017. So I wondered whether the idea I had about Prisma was workable. Turns out it is, but it involves a lot of hard work and I’m going to try and explain how I did it.
So it started with the video I shot on my iPhone 6S while I was at the fabric market looking for some lightweight material for the pants I needed. Mangaldas Market is a confusing place and we weren’t really sure what we were looking for. We got there pretty early–most of the shops weren’t even open yet–so as we were wandering around I saw this feline family. And then we had our raw material.
I made a black and white video which was straightforward and really not something I was too thrilled with. While the Prisma idea was percolating in the back of my mind I decided to sleep on it.
In the morning I brought back the color, and exported the video as a JPEG sequence. I stuck with the 25 FPS (frames per second) frame rate of the video which meant I had 225 individual frames to process in Prisma. I knew this was going to be difficult work. I started at around 7 AM this morning.
I decided to work with just the first 25 frames to see if I had something worth pursuing. It began with AirDropping the files from my desktop to my phone.
The first attempt was a disaster. I had exported 720p footage as is so it became very difficult for me, almost from the very beginning, to keep track of which frames I had processed in Prisma. When I felt confident that I had converted all the frames, I brought the footage back onto my computer (AirDrop is by far my favorite piece of technology right now) and imported the JPEG sequence into After Effects where I realized that the individual frames looked cool but the animation just did not work. I was ready to throw in the towel even thought it was not yet 9 AM. I even told myself that the black and white video I had already made was good enough. But then I decided to try again.
This time I decided to drop the 720p video into a 1080p Comp and render that out as a JPEG sequence with one crucial difference: this time my frames were individually numbered. And immediately I saw what the problem was while reviewing the footage on my phone. Even though I had AirDropped a sequence of 50 frames in numerical order, they were saved on my phone in any old order. Now I could have just processed each file as I saw it and worked on re-numbering all of them once they were back on the computer but I chose instead to make the effort to process them right there on my phone in the correct numerical order. This added hours to my effort but it was ultimately worth it.
Progress is Made
It was slow going at first and I had to go see the tailor for a trial fitting of my pants. Yes the story of my pants is hopelessly intertwined with how I used a smartphone app to make this video. On the car ride to and from the tailor’s—which took about two hours, all told—I had completed work on just 2 seconds of footage. When I was back at my computer though I decided to work on the material 25 frames at a time. As time went by and the film started looking like a reality I became better at remembering the order of the frames on my phone and so the processing time went by a lot faster than it had when I was staring at 225 unprocessed frames this morning.
I finally finished working on all the files around 3:30 PM and I am quite pleased with the way the video looks, but this was hard work. I know it wasn’t as difficult as actually animating each frame by hand but it was still a lot of effort for a video that is supposed to reside on Instagram, on an account that has less than 750 followers at this moment.
So Why Make The Effort?
When we agreed to undertake the challenge of making a video a day in 2017 we also promised that we would push ourselves in new creative directions. The first cut was good enough for us to honor our commitment of publishing a video a day but the new video is so much better. And I learnt how to hack Prisma to do what I had imagined doing with it all those months ago.
So, all told, it’s a good thing those pants split when they did.