4TH UG2+ Prize Challenge

The 4^th version of UG²+ Prize Challenge: CVPR 2021 will inherit the successful benchmark dataset, platform, and evaluation tools used by the first three UG² workshops but will also look at brand new aspects of the overall problem, significantly augmenting its existing scope. Sponsored by NVIDIA and Walmart, this workshop aims to promote video-based action recognition algorithms’ robustness with a special focus on dark videos, under both fully or semi-supervised action recognition in the dark.

Our heartiest congratulations to Team A*StarTrek for emerging as the winner with the highest accuracy for Track 2.1: Fully Supervised Action Recognition in the Dark.

A*StarTrek Team: 

• Institute of High Performance Computing (IHPC): Yang Zaifeng Yang
• Institute for Infocomm Research (I²R): Jin Ruibing, Wu Keyu, Hou Yubo, Chen Zhenghua, Wu Min