Abstract

First-person videos have unique characteristics such as heavy egocentric motion, strong preceding events, salient transitional activities and post-event impacts. Action recognition methods designed for third person videos may not optimally represent actions captured by fifirst-person videos. We propose a method to represent the high level dynamics of sub-events in fifirst-person videos by dynamically pooling features of sub-intervals of time series using a temporal feature pooling function. The sub-event dynamics are then temporally aligned to make a new series. To keep track of how the sub-event dynamics evolve over time, we recursively employ the Fast Fourier Transform on a pyramidal temporal structure. The Fourier coeffificients of the segment defifine the overall video representation. We perform experiments on two existing benchmark fifirst-person video datasets which have been captured in a controlled environment. Addressing this gap, we introduce a new dataset collected from YouTube which has a larger number of classes and a greater diversity of capture conditions thereby more closely depicting real-world challenges in fifirst-person video analysis. We compare our method to state-of-the-art fifirst person and generic video recognition algorithms. Our method consistently outperforms the nearest competitors by 10.3%, 3.3% and 11.7% respectively on the three datasets