NMS Threshold matters for Ego4D Moment Queries -- 2nd place solution to the Ego4D Moment Queries Challenge 2023
This work addresses video understanding for ego-centric datasets, but it is incremental as it builds on an existing method with refinements.
The paper tackled temporal action localization in the Ego4D Moment Queries Challenge by extending ActionFormer with improved ground-truth assignment and SoftNMS, achieving 26.62% average mAP and 45.69% Recall@1x at tIoU=0.5, placing 2nd on the leaderboard.
This report describes our submission to the Ego4D Moment Queries Challenge 2023. Our submission extends ActionFormer, a latest method for temporal action localization. Our extension combines an improved ground-truth assignment strategy during training and a refined version of SoftNMS at inference time. Our solution is ranked 2nd on the public leaderboard with 26.62% average mAP and 45.69% Recall@1x at tIoU=0.5 on the test set, significantly outperforming the strong baseline from 2023 challenge. Our code is available at https://github.com/happyharrycn/actionformer_release.