Loading…

SnapCap: Efficient Snapshot Compressive Video Captioning

Video Captioning (VC) is a challenging multi-modal task since it requires describing the scene in language by understanding various and complex videos. For machines, the traditional VC follows the "imaging-compression-decoding-and-then-captioning" pipeline, where compression is pivot for s...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2024-01
Main Authors: Sun, Jianqiao, Su, Yudi, Zhang, Hao, Cheng, Ziheng, Zeng, Zequn, Wang, Zhengjue, Chen, Bo, Yuan, Xin
Format: Article
Language:English
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Sun, Jianqiao
Su, Yudi
Zhang, Hao
Cheng, Ziheng
Zeng, Zequn
Wang, Zhengjue
Chen, Bo
Yuan, Xin
description Video Captioning (VC) is a challenging multi-modal task since it requires describing the scene in language by understanding various and complex videos. For machines, the traditional VC follows the "imaging-compression-decoding-and-then-captioning" pipeline, where compression is pivot for storage and transmission. However, in such a pipeline, some potential shortcomings are inevitable, i.e., information redundancy resulting in low efficiency and information loss during the sampling process for captioning. To address these problems, in this paper, we propose a novel VC pipeline to generate captions directly from the compressed measurement, which can be captured by a snapshot compressive sensing camera and we dub our model SnapCap. To be more specific, benefiting from the signal simulation, we have access to obtain abundant measurement-video-annotation data pairs for our model. Besides, to better extract language-related visual representations from the compressed measurement, we propose to distill the knowledge from videos via a pre-trained CLIP with plentiful language-vision associations to guide the learning of our SnapCap. To demonstrate the effectiveness of SnapCap, we conduct experiments on two widely-used VC datasets. Both the qualitative and quantitative results verify the superiority of our pipeline over conventional VC pipelines. In particular, compared to the "caption-after-reconstruction" methods, our SnapCap can run at least 3\(\times\) faster, and achieve better caption results.
format article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2913248818</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2913248818</sourcerecordid><originalsourceid>FETCH-proquest_journals_29132488183</originalsourceid><addsrcrecordid>eNqNitEKwiAUQCUIGrV_EHoe6HUr61UWvRe9DllajlLzur6_BX1ATwfOOTNSgBC8kjXAgpSIA2MMNltoGlEQefI6Kh33tLXW9c74TL8K7yFTFZ4xGUT3NvTiribQ6cwueOdvKzK3-oGm_HFJ1of2rI5VTOE1GszdEMbkp9TBjguopeRS_Hd9ADLPNgk</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2913248818</pqid></control><display><type>article</type><title>SnapCap: Efficient Snapshot Compressive Video Captioning</title><source>Publicly Available Content Database (Proquest) (PQ_SDU_P3)</source><creator>Sun, Jianqiao ; Su, Yudi ; Zhang, Hao ; Cheng, Ziheng ; Zeng, Zequn ; Wang, Zhengjue ; Chen, Bo ; Yuan, Xin</creator><creatorcontrib>Sun, Jianqiao ; Su, Yudi ; Zhang, Hao ; Cheng, Ziheng ; Zeng, Zequn ; Wang, Zhengjue ; Chen, Bo ; Yuan, Xin</creatorcontrib><description>Video Captioning (VC) is a challenging multi-modal task since it requires describing the scene in language by understanding various and complex videos. For machines, the traditional VC follows the "imaging-compression-decoding-and-then-captioning" pipeline, where compression is pivot for storage and transmission. However, in such a pipeline, some potential shortcomings are inevitable, i.e., information redundancy resulting in low efficiency and information loss during the sampling process for captioning. To address these problems, in this paper, we propose a novel VC pipeline to generate captions directly from the compressed measurement, which can be captured by a snapshot compressive sensing camera and we dub our model SnapCap. To be more specific, benefiting from the signal simulation, we have access to obtain abundant measurement-video-annotation data pairs for our model. Besides, to better extract language-related visual representations from the compressed measurement, we propose to distill the knowledge from videos via a pre-trained CLIP with plentiful language-vision associations to guide the learning of our SnapCap. To demonstrate the effectiveness of SnapCap, we conduct experiments on two widely-used VC datasets. Both the qualitative and quantitative results verify the superiority of our pipeline over conventional VC pipelines. In particular, compared to the "caption-after-reconstruction" methods, our SnapCap can run at least 3\(\times\) faster, and achieve better caption results.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Redundancy ; Video compression</subject><ispartof>arXiv.org, 2024-01</ispartof><rights>2024. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/2913248818?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>780,784,25753,37012,44590</link.rule.ids></links><search><creatorcontrib>Sun, Jianqiao</creatorcontrib><creatorcontrib>Su, Yudi</creatorcontrib><creatorcontrib>Zhang, Hao</creatorcontrib><creatorcontrib>Cheng, Ziheng</creatorcontrib><creatorcontrib>Zeng, Zequn</creatorcontrib><creatorcontrib>Wang, Zhengjue</creatorcontrib><creatorcontrib>Chen, Bo</creatorcontrib><creatorcontrib>Yuan, Xin</creatorcontrib><title>SnapCap: Efficient Snapshot Compressive Video Captioning</title><title>arXiv.org</title><description>Video Captioning (VC) is a challenging multi-modal task since it requires describing the scene in language by understanding various and complex videos. For machines, the traditional VC follows the "imaging-compression-decoding-and-then-captioning" pipeline, where compression is pivot for storage and transmission. However, in such a pipeline, some potential shortcomings are inevitable, i.e., information redundancy resulting in low efficiency and information loss during the sampling process for captioning. To address these problems, in this paper, we propose a novel VC pipeline to generate captions directly from the compressed measurement, which can be captured by a snapshot compressive sensing camera and we dub our model SnapCap. To be more specific, benefiting from the signal simulation, we have access to obtain abundant measurement-video-annotation data pairs for our model. Besides, to better extract language-related visual representations from the compressed measurement, we propose to distill the knowledge from videos via a pre-trained CLIP with plentiful language-vision associations to guide the learning of our SnapCap. To demonstrate the effectiveness of SnapCap, we conduct experiments on two widely-used VC datasets. Both the qualitative and quantitative results verify the superiority of our pipeline over conventional VC pipelines. In particular, compared to the "caption-after-reconstruction" methods, our SnapCap can run at least 3\(\times\) faster, and achieve better caption results.</description><subject>Redundancy</subject><subject>Video compression</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNqNitEKwiAUQCUIGrV_EHoe6HUr61UWvRe9DllajlLzur6_BX1ATwfOOTNSgBC8kjXAgpSIA2MMNltoGlEQefI6Kh33tLXW9c74TL8K7yFTFZ4xGUT3NvTiribQ6cwueOdvKzK3-oGm_HFJ1of2rI5VTOE1GszdEMbkp9TBjguopeRS_Hd9ADLPNgk</recordid><startdate>20240110</startdate><enddate>20240110</enddate><creator>Sun, Jianqiao</creator><creator>Su, Yudi</creator><creator>Zhang, Hao</creator><creator>Cheng, Ziheng</creator><creator>Zeng, Zequn</creator><creator>Wang, Zhengjue</creator><creator>Chen, Bo</creator><creator>Yuan, Xin</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20240110</creationdate><title>SnapCap: Efficient Snapshot Compressive Video Captioning</title><author>Sun, Jianqiao ; Su, Yudi ; Zhang, Hao ; Cheng, Ziheng ; Zeng, Zequn ; Wang, Zhengjue ; Chen, Bo ; Yuan, Xin</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_29132488183</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Redundancy</topic><topic>Video compression</topic><toplevel>online_resources</toplevel><creatorcontrib>Sun, Jianqiao</creatorcontrib><creatorcontrib>Su, Yudi</creatorcontrib><creatorcontrib>Zhang, Hao</creatorcontrib><creatorcontrib>Cheng, Ziheng</creatorcontrib><creatorcontrib>Zeng, Zequn</creatorcontrib><creatorcontrib>Wang, Zhengjue</creatorcontrib><creatorcontrib>Chen, Bo</creatorcontrib><creatorcontrib>Yuan, Xin</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>AUTh Library subscriptions: ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection (Proquest) (PQ_SDU_P3)</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database (Proquest) (PQ_SDU_P3)</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Sun, Jianqiao</au><au>Su, Yudi</au><au>Zhang, Hao</au><au>Cheng, Ziheng</au><au>Zeng, Zequn</au><au>Wang, Zhengjue</au><au>Chen, Bo</au><au>Yuan, Xin</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>SnapCap: Efficient Snapshot Compressive Video Captioning</atitle><jtitle>arXiv.org</jtitle><date>2024-01-10</date><risdate>2024</risdate><eissn>2331-8422</eissn><abstract>Video Captioning (VC) is a challenging multi-modal task since it requires describing the scene in language by understanding various and complex videos. For machines, the traditional VC follows the "imaging-compression-decoding-and-then-captioning" pipeline, where compression is pivot for storage and transmission. However, in such a pipeline, some potential shortcomings are inevitable, i.e., information redundancy resulting in low efficiency and information loss during the sampling process for captioning. To address these problems, in this paper, we propose a novel VC pipeline to generate captions directly from the compressed measurement, which can be captured by a snapshot compressive sensing camera and we dub our model SnapCap. To be more specific, benefiting from the signal simulation, we have access to obtain abundant measurement-video-annotation data pairs for our model. Besides, to better extract language-related visual representations from the compressed measurement, we propose to distill the knowledge from videos via a pre-trained CLIP with plentiful language-vision associations to guide the learning of our SnapCap. To demonstrate the effectiveness of SnapCap, we conduct experiments on two widely-used VC datasets. Both the qualitative and quantitative results verify the superiority of our pipeline over conventional VC pipelines. In particular, compared to the "caption-after-reconstruction" methods, our SnapCap can run at least 3\(\times\) faster, and achieve better caption results.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2024-01
issn 2331-8422
language eng
recordid cdi_proquest_journals_2913248818
source Publicly Available Content Database (Proquest) (PQ_SDU_P3)
subjects Redundancy
Video compression
title SnapCap: Efficient Snapshot Compressive Video Captioning
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T22%3A20%3A45IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=SnapCap:%20Efficient%20Snapshot%20Compressive%20Video%20Captioning&rft.jtitle=arXiv.org&rft.au=Sun,%20Jianqiao&rft.date=2024-01-10&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2913248818%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-proquest_journals_29132488183%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2913248818&rft_id=info:pmid/&rfr_iscdi=true