Building Resilient Vision Systems

Consequently, temporal prediction has been established as a way to achieve a deeper understanding of the data as it requires an implicit understanding of the structure of the observed visual features and the rules they follow while they change over time. Given this, recent approaches in the ?eld usually extract the temporal properties of the data by learning a strong temporal regularization [85, 147], which helps to ?nd clusters over time but may result in a lower ability to identify segments based on their visual representation.