Show simple item record

dc.contributor.authorGiancola, Silvio
dc.contributor.authorGhanem, Bernard
dc.date.accessioned2021-10-06T05:27:13Z
dc.date.available2021-04-19T08:59:13Z
dc.date.available2021-10-06T05:27:13Z
dc.date.issued2021-06
dc.identifier.citationGiancola, S., & Ghanem, B. (2021). Temporally-Aware Feature Pooling for Action Spotting in Soccer Broadcasts. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). doi:10.1109/cvprw53098.2021.00506
dc.identifier.doi10.1109/cvprw53098.2021.00506
dc.identifier.urihttp://hdl.handle.net/10754/668843
dc.description.abstractToward the goal of automatic production for sports broadcasts, a paramount task consists in understanding the high-level semantic information of the game in play. For instance, recognizing and localizing the main actions of the game would allow producers to adapt and automatize the broadcast production, focusing on the important details of the game and maximizing the spectator engagement. In this paper, we focus our analysis on action spotting in soccer broadcast, which consists in temporally localizing the main actions in a soccer game. To that end, we propose a novel feature pooling method based on NetVLAD, dubbed NetVLAD ++ , that embeds temporally-aware knowledge. Different from previous pooling methods that consider the temporal context as a single set to pool from, we split the context before and after an action occurs. We argue that considering the contextual information around the action spot as a single entity leads to a sub-optimal learning for the pooling module. With NetVLAD ++ , we disentangle the context from the past and future frames and learn specific vocabularies of semantics for each subsets, avoiding to blend and blur such vocabulary in time. Injecting such prior knowledge creates more informative pooling modules and more discriminative pooled features, leading into a better understanding of the actions. We train and evaluate our methodology on the recent large-scale dataset SoccerNet-v2, reaching 53.4% Average-mAP for action spotting, a +12.7% improvement w.r.t the current state-of-the-art.
dc.description.sponsorshipThis work is supported by the KAUST Office of Sponsored Research under Award No. OSR-CRG2017-3405.
dc.publisherIEEE
dc.relation.urlhttps://ieeexplore.ieee.org/document/9523061/
dc.rightsArchived with thanks to IEEE
dc.titleTemporally-Aware Feature Pooling for Action Spotting in Soccer Broadcasts
dc.typeConference Paper
dc.contributor.departmentComputer, Electrical and Mathematical Science and Engineering (CEMSE) Division
dc.contributor.departmentElectrical and Computer Engineering Program
dc.contributor.departmentVCC Analytics Research Group
dc.contributor.departmentVisual Computing Center (VCC)
dc.conference.date19-25 June 2021
dc.conference.name2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops
dc.conference.locationNashville, TN, USA
dc.eprint.versionPost-print
dc.identifier.arxivid2104.06779
kaust.personGiancola, Silvio
kaust.personGhanem, Bernard
kaust.grant.numberOSR-CRG2017-3405
refterms.dateFOA2021-04-19T08:59:45Z
kaust.acknowledged.supportUnitCRG
kaust.acknowledged.supportUnitKAUST Office of Sponsored Research
kaust.acknowledged.supportUnitOSR
dc.date.posted2021-04-14


Files in this item

Thumbnail
Name:
Preprintfile1.pdf
Size:
1.434Mb
Format:
PDF
Description:
Pre-print

This item appears in the following Collection(s)

Show simple item record

VersionItemEditorDateSummary

*Selected version