The Wisdom of Crowds: Temporal Progressive Attention for Early Action Prediction

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

Early action prediction deals with inferring the ongoing action from partially-observed videos, typically at the outset of the video. We propose a bottleneck-based attention model that captures the evolution of the action, through progressive sampling over fine-to-coarse scales. Our proposed Temporal Progressive (TemPr) model is composed of multiple attention towers, one for each scale. The predicted action label is based on the collective agreement considering confidences of these towers. Extensive experiments over four video datasets showcase state-of-the-art performance on the task of Early Action Prediction across a range of encoder architectures. We demonstrate the effectiveness and consistency of TemPr through detailed ablations. † † Code is available at: https://tinyurl.com/temprog
Original languageEnglish
Title of host publication2023 IEEE/CVF Conference On Computer Vision And Pattern Recognition
Subtitle of host publicationCVPR 2023
PublisherIEEE
Pages14709-14719
Number of pages11
ISBN (Print)979-8-3503-0129-8
DOIs
Publication statusPublished - 2023
Externally publishedYes
EventIEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023 - Vancouver, Canada
Duration: 18 Jun 202322 Jun 2023
https://cvpr2023.thecvf.com/

Publication series

NameProceedings
PublisherIEEE
ISSN (Print)2575-7075

Conference

ConferenceIEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023
Abbreviated titleCVPR 2023
Country/TerritoryCanada
CityVancouver
Period18/06/2322/06/23
Internet address

Keywords

  • n/a OA procedure

Fingerprint

Dive into the research topics of 'The Wisdom of Crowds: Temporal Progressive Attention for Early Action Prediction'. Together they form a unique fingerprint.

Cite this