TY - GEN
T1 - RefineLoc: Iterative Refinement for Weakly-Supervised Action Localization
AU - Pardo, Alejandro
AU - Alwassel, Humam
AU - Heilbron, Fabian Caba
AU - Thabet, Ali Kassem
AU - Ghanem, Bernard
N1 - KAUST Repository Item: Exported on 2021-06-17
Acknowledged KAUST grant number(s): OSRCRG2017-3405
Acknowledgements: This work is supported the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research (OSR) under Award No. OSRCRG2017-3405.
PY - 2021
Y1 - 2021
N2 - Video action detectors are usually trained using datasets with fully-supervised temporal annotations. Building such datasets is an expensive task. To alleviate this problem, recent methods have tried to leverage weak labeling, where videos are untrimmed and only a video-level label is available. In this paper, we propose RefineLoc, a novel weaklysupervised temporal action localization method. RefineLoc uses an iterative refinement approach by estimating and training on snippet-level pseudo ground truth at every iteration. We show the benefit of this iterative approach and present an extensive analysis of five different pseudo ground truth generators. We show the effectiveness of our model on two standard action datasets, ActivityNet v1.2 and THUMOS14. RefineLoc shows competitive results with the stateof-the-art in weakly-supervised temporal localization. Additionally, our iterative refinement process is able to significantly improve the performance of two state-of-the-art methods, setting a new state-of-the-art on THUMOS14.
AB - Video action detectors are usually trained using datasets with fully-supervised temporal annotations. Building such datasets is an expensive task. To alleviate this problem, recent methods have tried to leverage weak labeling, where videos are untrimmed and only a video-level label is available. In this paper, we propose RefineLoc, a novel weaklysupervised temporal action localization method. RefineLoc uses an iterative refinement approach by estimating and training on snippet-level pseudo ground truth at every iteration. We show the benefit of this iterative approach and present an extensive analysis of five different pseudo ground truth generators. We show the effectiveness of our model on two standard action datasets, ActivityNet v1.2 and THUMOS14. RefineLoc shows competitive results with the stateof-the-art in weakly-supervised temporal localization. Additionally, our iterative refinement process is able to significantly improve the performance of two state-of-the-art methods, setting a new state-of-the-art on THUMOS14.
UR - http://hdl.handle.net/10754/660670
UR - https://ieeexplore.ieee.org/document/9423165/
U2 - 10.1109/WACV48630.2021.00336
DO - 10.1109/WACV48630.2021.00336
M3 - Conference contribution
SN - 978-1-6654-4640-2
BT - 2021 IEEE Winter Conference on Applications of Computer Vision (WACV)
PB - IEEE
ER -