Evolution-Preserving Dense Trajectory Descriptors

TL;DR

This paper introduces Evolution-Preserving Trajectory (EPT) descriptors, which encode the temporal evolution of deep features along trajectories, significantly improving human action recognition performance.

Contribution

The paper proposes a novel EPT descriptor that applies rank pooling to dense trajectories, enhancing the encoding of temporal evolution in video analysis.

Findings

EPT descriptors outperform previous trajectory-pooled deep descriptors

Combining EPT with VideoDarwin achieves state-of-the-art results

EPT provides complementary benefits to non-trajectory-based descriptors

Abstract

Recently Trajectory-pooled Deep-learning Descriptors were shown to achieve state-of-the-art human action recognition results on a number of datasets. This paper improves their performance by applying rank pooling to each trajectory, encoding the temporal evolution of deep learning features computed along the trajectory. This leads to Evolution-Preserving Trajectory (EPT) descriptors, a novel type of video descriptor that significantly outperforms Trajectory-pooled Deep-learning Descriptors. EPT descriptors are defined based on dense trajectories, and they provide complimentary benefits to video descriptors that are not based on trajectories. In particular, we show that the combination of EPT descriptors and VideoDarwin leads to state-of-the-art performance on Hollywood2 and UCF101 datasets.