Graph Regularization Network with Semantic Affinity for Weakly-Supervised Temporal Action Localization
- Authors
- Park, Jungin; Lee, Jiyoung; Jeon, Sangryul; Kim, Seungryong; Sohn, Kwanghoon
- Issue Date
- Sep-2019
- Publisher
- IEEE Computer Society
- Keywords
- graph Laplacian regularization; semantic affinity; weakly-supervised temporal action localization
- Citation
- Proceedings - International Conference on Image Processing, ICIP, v.2019-September, pp 3701 - 3705
- Pages
- 5
- Journal Title
- Proceedings - International Conference on Image Processing, ICIP
- Volume
- 2019-September
- Start Page
- 3701
- End Page
- 3705
- URI
- https://yscholarhub.yonsei.ac.kr/handle/2021.sw.yonsei/6620
- DOI
- 10.1109/ICIP.2019.8803589
- ISSN
- 1522-4880
- Abstract
- This paper presents a novel deep architecture for weakly-supervised temporal action localization that not only generates segment-level action responses but also propagates segment-level responses to the neighborhood in a form of graph Laplacian regularization. Specifically, our approach consists of two sub-modules; a class activation module to estimate the action score map over time through the action classifiers, and a graph regularization module to refine the estimated action score map by solving a quadratic programming problem with the predicted segment-level semantic affinities. Since these two modules are integrated with fully differentiable layers, the proposed networks can be jointly trained in an end-to-end manner. Experimental results on Thumos14 and ActivityNet1.2 demonstrate that the proposed method provides outstanding performances in weakly-supervised temporal action localization. © 2019 IEEE.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - College of Engineering > Electrical and Electronic Engineering > 1. Journal Articles
Items in Scholar Hub are protected by copyright, with all rights reserved, unless otherwise indicated.