Home

Self-Supervised Action Recognition In The Dark (s2arid)

Overview

Action Recognition (AR) has gained large improvements with the introduction of large-scale video datasets and the development of deep neural networks. However, AR models robust to challenging environments in real-world scenarios are still under-explored. We focus on the task of video-based action recognition in dark environments, which can be applied to fields such as surveillance and autonomous driving at night. Through this project, we intend to participate in the UG2+ Challenge Track 2 UG2-2 which evaluates the robustness of action recognition methods in dark environments.