Where a Strong Backbone Meets Strong Features – ActionFormer for Ego4D Moment Queries Challenge

11/16/2022

∙

by   Fangzhou Mu, et al.

∙

0

∙

share

This report describes our submission to the Ego4D Moment Queries Challenge 2022. Our submission builds on ActionFormer, the state-of-the-art backbone for temporal action localization, and a trio of strong video features from SlowFast, Omnivore and EgoVLP. Our solution is ranked 2nd on the public leaderboard with 21.76 times higher than the official baseline. Further, we obtain 42.54 tIoU=0.5 on the test set, outperforming the top-ranked solution by a significant margin of 1.41 absolute percentage points. Our code is available at https://github.com/happyharrycn/actionformer_release.