Inspiration
machine learning is the future. why make videos when you can just take two pictures and have the computer generate a video for you?
What it does
takes two images and creates a model interpolating between them, then use that model to predictively generate future frames
How we built it
numpy for linear algebra and pillow for image processing
Challenges we ran into
this thing is single threaded and will take literally forever on larger/longer frames
Accomplishments that we're proud of
it actually works
What we learned
linear models make nice colors
What's next for the extrapolator
polynomial fits with more source images
Log in or sign up for Devpost to join the conversation.