Show simple item record

dc.contributor.authorEscorcia, Victor
dc.contributor.authorDao, Cuong D.
dc.contributor.authorJain, Mihir
dc.contributor.authorGhanem, Bernard
dc.contributor.authorSnoek, Cees
dc.date.accessioned2018-04-16T11:27:42Z
dc.date.available2018-04-16T11:27:42Z
dc.date.issued2018-04-05
dc.identifier.urihttp://hdl.handle.net/10754/627508
dc.description.abstractThis paper addresses the problem of spatiotemporal localization of actions in videos. Compared to leading approaches, which all learn to localize based on carefully annotated boxes on training video frames, we adhere to a weakly-supervised solution that only requires a video class label. We introduce an actor-supervised architecture that exploits the inherent compositionality of actions in terms of actor transformations, to localize actions. We make two contributions. First, we propose actor proposals derived from a detector for human and non-human actors intended for images, which is linked over time by Siamese similarity matching to account for actor deformations. Second, we propose an actor-based attention mechanism that enables the localization of the actions from action class labels and actor proposals and is end-to-end trainable. Experiments on three human and non-human action datasets show actor supervision is state-of-the-art for weakly-supervised action localization and is even competitive to some fully-supervised alternatives.
dc.publisherarXiv
dc.relation.urlhttp://arxiv.org/abs/1804.01824v1
dc.relation.urlhttp://arxiv.org/pdf/1804.01824v1
dc.rightsArchived with thanks to arXiv
dc.titleGuess Where? Actor-Supervision for Spatiotemporal Action Localization
dc.typePreprint
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.contributor.departmentElectrical Engineering Program
dc.contributor.departmentVisual Computing Center (VCC)
dc.eprint.versionPre-print
dc.contributor.institutionQualcomm Technologies, Inc.
dc.contributor.institutionUniversity of Amsterdam
dc.identifier.arxividarXiv:1804.01824
kaust.personEscorcia, Victor
kaust.personDao, Cuong D.
kaust.personGhanem, Bernard
refterms.dateFOA2018-06-14T04:20:48Z


Files in this item

Thumbnail
Name:
1804.01824v1.pdf
Size:
2.774Mb
Format:
PDF
Description:
Preprint

This item appears in the following Collection(s)

Show simple item record