<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="pirhadi-etal-2025-cvt5">
    <titleInfo>
        <title>CVT5: Using Compressed Video Encoder and UMT5 for Dense Video Captioning</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Mohammad</namePart>
        <namePart type="given">Javad</namePart>
        <namePart type="family">Pirhadi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Motahhare</namePart>
        <namePart type="family">Mirzaei</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Sauleh</namePart>
        <namePart type="family">Eetemadi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-01</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the First Workshop of Evaluation of Multi-Modal Generation</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Wei</namePart>
            <namePart type="given">Emma</namePart>
            <namePart type="family">Zhang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Xiang</namePart>
            <namePart type="family">Dai</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Desmond</namePart>
            <namePart type="family">Elliot</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Byron</namePart>
            <namePart type="family">Fang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Mongyuan</namePart>
            <namePart type="family">Sim</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Haojie</namePart>
            <namePart type="family">Zhuang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Weitong</namePart>
            <namePart type="family">Chen</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Abu Dhabi, UAE</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>The dense video captioning task aims to detect all events occurring in a video and describe each event using natural language. Unlike most other video processing tasks, where it is typically assumed that videos contain only a single main event, this task deals with long, untrimmed videos. Consequently, the speed of processing videos in dense video captioning is a critical aspect of the system. To the best of our knowledge, all published work on this task uses RGB frames to encode input videos. In this work, we introduce the use of compressed videos for the first time in this task. Our experiments on the SoccerNet challenge demonstrate significant improvements in both processing speed and GPU memory footprint while achieving competitive results. Additionally, we leverage multilingual transcripts, which seems to be effective. The encoder in our proposed method achieves approximately 5.4× higher speed and 5.1× lower GPU memory usage during training, and 4.7× higher speed and 7.8× lower GPU memory usage during inference, compared to its RGB-based counterpart. The code is publicly available at https://github.com/mohammadjavadpirhadi/CVT5.</abstract>
    <identifier type="citekey">pirhadi-etal-2025-cvt5</identifier>
    <location>
        <url>https://aclanthology.org/2025.evalmg-1.2/</url>
    </location>
    <part>
        <date>2025-01</date>
        <extent unit="page">
            <start>10</start>
            <end>23</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T CVT5: Using Compressed Video Encoder and UMT5 for Dense Video Captioning
%A Pirhadi, Mohammad Javad
%A Mirzaei, Motahhare
%A Eetemadi, Sauleh
%Y Zhang, Wei Emma
%Y Dai, Xiang
%Y Elliot, Desmond
%Y Fang, Byron
%Y Sim, Mongyuan
%Y Zhuang, Haojie
%Y Chen, Weitong
%S Proceedings of the First Workshop of Evaluation of Multi-Modal Generation
%D 2025
%8 January
%I Association for Computational Linguistics
%C Abu Dhabi, UAE
%F pirhadi-etal-2025-cvt5
%X The dense video captioning task aims to detect all events occurring in a video and describe each event using natural language. Unlike most other video processing tasks, where it is typically assumed that videos contain only a single main event, this task deals with long, untrimmed videos. Consequently, the speed of processing videos in dense video captioning is a critical aspect of the system. To the best of our knowledge, all published work on this task uses RGB frames to encode input videos. In this work, we introduce the use of compressed videos for the first time in this task. Our experiments on the SoccerNet challenge demonstrate significant improvements in both processing speed and GPU memory footprint while achieving competitive results. Additionally, we leverage multilingual transcripts, which seems to be effective. The encoder in our proposed method achieves approximately 5.4× higher speed and 5.1× lower GPU memory usage during training, and 4.7× higher speed and 7.8× lower GPU memory usage during inference, compared to its RGB-based counterpart. The code is publicly available at https://github.com/mohammadjavadpirhadi/CVT5.
%U https://aclanthology.org/2025.evalmg-1.2/
%P 10-23

Markdown (Informal)

[CVT5: Using Compressed Video Encoder and UMT5 for Dense Video Captioning](https://aclanthology.org/2025.evalmg-1.2/) (Pirhadi et al., EvalMG 2025)

• CVT5: Using Compressed Video Encoder and UMT5 for Dense Video Captioning (Pirhadi et al., EvalMG 2025)

ACL

• Mohammad Javad Pirhadi, Motahhare Mirzaei, and Sauleh Eetemadi. 2025. CVT5: Using Compressed Video Encoder and UMT5 for Dense Video Captioning. In Proceedings of the First Workshop of Evaluation of Multi-Modal Generation, pages 10–23, Abu Dhabi, UAE. Association for Computational Linguistics.