Speeding up inference times for our custom first order motion model for low bandwidth video conferencing. We want to explore sparsification, quantization, and target architecture specific techniques to speed up our model inference time.

I want to explore research in a structured way and work on a project from start to end. I have a little bit of background on this specific project from the past two years and I’d like to explore fast inference in machine learning in depth, so this seems like the perfect project.