University of Waterloo

Training Pose Models with Few Annotations

Brennan Gebotys

Apr 1, 2022, 11:30am EC4-2101A

With the recent advances in machine learning, a field likely to benefit greatly is video analytics: the analysis of video data. Two applications of importance include pose estimation and action recognition. However, key problems such as how to train a pose estimation model with a small number of annotations and how to design an action recognition model to achieve the highest possible accuracy still remain. This presentation explores how effectively leveraging motion information can enable strategies that can solve both of these problems.