We address the problem of extracting key steps from unlabeled procedural...
We address the problem of few-shot classification where the goal is to l...
We present a solution for the goal of extracting a video from a single m...
Standard contrastive learning approaches usually require a large number ...
A core component of the recent success of self-supervised learning is
cr...
Recent literature has shown that features obtained from supervised train...
Most human action recognition systems typically consider static appearan...
We present a solution for the novel goal of extracting a video from a si...