Notice

This is not the latest version of this item. The latest version can be found at: https://repository.kaust.edu.sa/handle/10754/660670

Show simple item record

dc.contributor.authorAlwassel, Humam
dc.contributor.authorHeilbron, Fabian Caba
dc.contributor.authorThabet, Ali
dc.contributor.authorGhanem, Bernard
dc.date.accessioned2019-12-18T11:23:14Z
dc.date.available2019-12-18T11:23:14Z
dc.date.issued2019-03-30
dc.identifier.urihttp://hdl.handle.net/10754/660670.1
dc.description.abstractVideo action detectors are usually trained using video datasets with fully supervised temporal annotations. Building such video datasets is a heavily expensive task. To alleviate this problem, recent algorithms leverage weak labelling where videos are untrimmed and only a video-level label is available. In this paper, we propose RefineLoc, a new method for weakly-supervised temporal action localization. RefineLoc uses an iterative refinement approach by estimating and training on snippet-level pseudo ground truth at every iteration. We show the benefit of using such an iterative approach and present an extensive analysis of different pseudo ground truth generators. We show the effectiveness of our model on two standard action datasets, ActivityNet v1.2 and THUMOS14. RefineLoc equipped with a segment prediction-based pseudo ground truth generator improves the state-of-the-art in weakly-supervised temporal localization on the challenging and large-scale ActivityNet dataset by 4.2% and achieves comparable performance with state-of-the-art on THUMOS14.
dc.publisherarXiv
dc.relation.urlhttps://arxiv.org/pdf/1904.00227
dc.rightsArchived with thanks to arXiv
dc.titleRefineLoc: Iterative Refinement for Weakly-Supervised Action Localization
dc.typePreprint
dc.eprint.versionPre-print
dc.identifier.arxivid1904.00227
refterms.dateFOA2019-12-18T11:24:16Z


Files in this item

Thumbnail
Name:
Preprintfile1.pdf
Size:
5.129Mb
Format:
PDF
Description:
Pre-print

This item appears in the following Collection(s)

Show simple item record

VersionItemEditorDateSummary

*Selected version