Extending Temporal Data Augmentation for Video Action Recognition

TL;DR

This paper introduces advanced temporal data augmentation techniques for video action recognition, improving performance by better integrating spatial and temporal perturbations, and demonstrating superior results on benchmark datasets.

Contribution

It proposes novel enhancements to temporal data augmentation methods, strengthening spatial-temporal relationships and achieving better recognition accuracy.

Findings

Outperforms existing augmentation variants on UCF-101 and HMDB-51 datasets.

Achieves higher Top-1 and Top-5 accuracy in video action recognition.

Demonstrates the effectiveness of temporal augmentations over spatial-only methods.

Abstract

Pixel space augmentation has grown in popularity in many Deep Learning areas, due to its effectiveness, simplicity, and low computational cost. Data augmentation for videos, however, still remains an under-explored research topic, as most works have been treating inputs as stacks of static images rather than temporally linked series of data. Recently, it has been shown that involving the time dimension when designing augmentations can be superior to its spatial-only variants for video action recognition. In this paper, we propose several novel enhancements to these techniques to strengthen the relationship between the spatial and temporal domains and achieve a deeper level of perturbations. The video action recognition results of our techniques outperform their respective variants in Top-1 and Top-5 settings on the UCF-101 and the HMDB-51 datasets.