Show simple item record

dc.contributor.authorGao, Jialin
dc.contributor.authorSun, Xin
dc.contributor.authorGhanem, Bernard
dc.contributor.authorZhou, Xi
dc.contributor.authorGe, Shiming
dc.date.accessioned2022-05-11T06:23:38Z
dc.date.available2022-05-11T06:23:38Z
dc.date.issued2022-05-10
dc.identifier.citationGao, J., Sun, X., Ghanem, B., Zhou, X., & Ge, S. (2022). Efficient Video Grounding with Which-Where Reading Comprehension. IEEE Transactions on Circuits and Systems for Video Technology, 1–1. https://doi.org/10.1109/tcsvt.2022.3174136
dc.identifier.issn1051-8215
dc.identifier.issn1558-2205
dc.identifier.doi10.1109/tcsvt.2022.3174136
dc.identifier.urihttp://hdl.handle.net/10754/676735
dc.description.abstractVideo grounding aims at localizing the temporal moment related to the given language description, which is very helpful to many cross-modal content understanding applications like visual question answering and sentence-video search. Existing approaches usually directly regress the temporal boundaries of an event described by a query sentence in the video sequence. This direct regression manner often encounters a large decision space due to diverse target events and variable video durations, leading to inaccurate localization as well as inefficient grounding. This paper presents an efficient framework termed from which to where to facilitate video grounding. The core idea is imitating the reading comprehension process to gradually narrow the decision space, in what we decompose the direct regression into two steps. The “which" step first roughly selects a candidate area by evaluating which video segment in the predefined set is closest to the ground truth. To this end, we formulate this step into a multi-choice reading comprehension problem and propose a criterion to select the best-matched segment. In this way, the excessive decision space is effectively reduced. The “where" step aims to precisely regress the temporal boundary of the selected video segment from the shrunk decision space. We thus introduce a triple-span representation for each candidate video segment to use the regional context for better boundary regression. The “which" and “where" steps can be combined into a unified framework and learned end-to-end, leading to an efficient video grounding system. Extensive experiments on Charades-STA, ActivityNet-Captions, and TACoS benchmarks clearly demonstrate the effectiveness of our framework.
dc.description.sponsorshipSupported by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research through the Visual Computing Center (VCC) funding, the Beijing Natural Science Foundation (19L2040) and the National Natural Science Foundation of China (61772513). We also thank the support from CloudWalk Technology Co., Ltd.
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)
dc.relation.urlhttps://ieeexplore.ieee.org/document/9771472/
dc.rights(c) 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.
dc.titleEfficient Video Grounding with Which-Where Reading Comprehension
dc.typeArticle
dc.contributor.departmentElectrical and Computer Engineering Program
dc.contributor.departmentComputer, Electrical and Mathematical Science and Engineering (CEMSE) Division
dc.contributor.departmentVisual Computing Center (VCC)
dc.identifier.journalIEEE Transactions on Circuits and Systems for Video Technology
dc.eprint.versionPost-print
dc.contributor.institutionCooperative Medianet Innovation Center, Shanghai Jiao Tong University, Shanghai, 200240, China.
dc.contributor.institutionCloudWalk Technology, Shanghai, 200240, China.
dc.contributor.institutionInstitute of Information Engineering, Chinese Academy of Sciences, Beijing 100095, China, and the School of Cyber Security, University of Chinese Academy of Sciences, Beijing 100049, China.
dc.identifier.pages1-1
kaust.personGhanem, Bernard
refterms.dateFOA2022-05-11T06:24:53Z
kaust.acknowledged.supportUnitOffice of Sponsored Research
kaust.acknowledged.supportUnitVisual Computing Center (VCC) funding


Files in this item

Thumbnail
Name:
Efficient_Video_Grounding_with_Which-Where_Reading_Comprehension.pdf
Size:
5.873Mb
Format:
PDF
Description:
Accepted Manuscript

This item appears in the following Collection(s)

Show simple item record