Going Deeper into Recognizing Actions in Dark Environments: A Comprehensive Benchmark Study

TL;DR

This paper investigates the challenges of action recognition in dark environments, introduces a new dataset and benchmark, and evaluates current models' performance, highlighting significant room for improvement in real-world scenarios.

Contribution

It presents the UG2+ Challenge Track 2 and the ARID dataset, providing a comprehensive benchmark for dark environment action recognition and fostering research in this under-explored area.

Findings

Current models show limited robustness in dark environments.

Active research participation has led to notable advances.

Analysis suggests promising directions for future improvements.

Abstract

While action recognition (AR) has gained large improvements with the introduction of large-scale video datasets and the development of deep neural networks, AR models robust to challenging environments in real-world scenarios are still under-explored. We focus on the task of action recognition in dark environments, which can be applied to fields such as surveillance and autonomous driving at night. Intuitively, current deep networks along with visual enhancement techniques should be able to handle AR in dark environments, however, it is observed that this is not always the case in practice. To dive deeper into exploring solutions for AR in dark environments, we launched the UG2+ Challenge Track 2 (UG2-2) in IEEE CVPR 2021, with a goal of evaluating and advancing the robustness of AR models in dark environments. The challenge builds and expands on top of a novel ARID dataset, the first dataset for the task of dark video AR, and guides models to tackle such a task in both fully and semi-supervised manners. Baseline results utilizing current AR models and enhancement methods are reported, justifying the challenging nature of this task with substantial room for improvements. Thanks to the active participation from the research community, notable advances have been made in participants' solutions, while analysis of these solutions helped better identify possible directions to tackle the challenge of AR in dark environments.