ITE Technical Group Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2026-02-19 15:45
A Note on Automated Dataset Construction for Composed Video Retrieval in Soccer -- Improvement of MLLM-based Dataset Evaluation via Prompt Tuning --
Riku Yoshida, Ryota Goka, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama (Hokkaido Univ.)
Abstract (in Japanese) (See Japanese page) 
(in English) In this paper, we investigate the automatic construction of a dataset for composed video retrieval (CoVR) targeting soccer videos. With the widespread adoption of sports video streaming services in recent years, there is a growing demand to efficiently retrieve and analyze specific plays or scenes from match videos. Combining query video with textual modification instructions for retrieval, CoVR is capable of flexibly reflecting user intent, and its application in the sports domain is also expected. CoVR requires triplets consisting of a query video, an instruction text, and a target video. However, existing automatic construction methods struggle with ``noise" caused by misalignment between instructions and video differences. This issue is particularly pronounced in soccer videos, where multiple players are involved, and situations change in complex ways. To address this, we automatically construct triplets using action/result labels summarized from live commentary captions and video features pre-trained on soccer videos. Furthermore, we employ a multimodal large language model (MLLM) enhanced with prompt tuning to automatically evaluate the relationships among the components of a triplet. At the end, we demonstrate the effectiveness of our proposed framework through experiments comparing MLLM evaluation with human evaluation.
Keyword (in Japanese) (See Japanese page) 
(in English) Composed video retrieval / triplet construction / prompt tuning / dataset evaluation / / / /  
Reference Info. ITE Tech. Rep., vol. 50, no. 5, ME2026-16, pp. 70-74, Feb. 2026.
Paper # ME2026-16 
Date of Issue 2026-02-12 (MMS, ME, AIT, SIP) 
ISSN Online edition: ISSN 2424-1970
Download PDF

Conference Information
Committee IEICE-IE IEICE-ITS MMS ME AIT SIP  
Conference Date 2026-02-19 - 2026-02-20 
Place (in Japanese) (See Japanese page) 
Place (in English)  
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To ME 
Conference Code 2026-02-IE-ITS-MMS-ME-AIT-SIP 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) A Note on Automated Dataset Construction for Composed Video Retrieval in Soccer 
Sub Title (in English) Improvement of MLLM-based Dataset Evaluation via Prompt Tuning 
Keyword(1) Composed video retrieval  
Keyword(2) triplet construction  
Keyword(3) prompt tuning  
Keyword(4) dataset evaluation  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Riku Yoshida  
1st Author's Affiliation Hokkaido University (Hokkaido Univ.)
2nd Author's Name Ryota Goka  
2nd Author's Affiliation Hokkaido University (Hokkaido Univ.)
3rd Author's Name Keisuke Maeda  
3rd Author's Affiliation Hokkaido University (Hokkaido Univ.)
4th Author's Name Takahiro Ogawa  
4th Author's Affiliation Hokkaido University (Hokkaido Univ.)
5th Author's Name Miki Haseyama  
5th Author's Affiliation Hokkaido University (Hokkaido Univ.)
6th Author's Name  
6th Author's Affiliation ()
7th Author's Name  
7th Author's Affiliation ()
8th Author's Name  
8th Author's Affiliation ()
9th Author's Name  
9th Author's Affiliation ()
10th Author's Name  
10th Author's Affiliation ()
11th Author's Name  
11th Author's Affiliation ()
12th Author's Name  
12th Author's Affiliation ()
13th Author's Name  
13th Author's Affiliation ()
14th Author's Name  
14th Author's Affiliation ()
15th Author's Name  
15th Author's Affiliation ()
16th Author's Name  
16th Author's Affiliation ()
17th Author's Name  
17th Author's Affiliation ()
18th Author's Name  
18th Author's Affiliation ()
19th Author's Name  
19th Author's Affiliation ()
20th Author's Name  
20th Author's Affiliation ()
21st Author's Name  
21st Author's Affiliation ()
22nd Author's Name  
22nd Author's Affiliation ()
23rd Author's Name  
23rd Author's Affiliation ()
24th Author's Name  
24th Author's Affiliation ()
25th Author's Name  
25th Author's Affiliation ()
26th Author's Name / /
26th Author's Affiliation ()
()
27th Author's Name / /
27th Author's Affiliation ()
()
28th Author's Name / /
28th Author's Affiliation ()
()
29th Author's Name / /
29th Author's Affiliation ()
()
30th Author's Name / /
30th Author's Affiliation ()
()
31st Author's Name / /
31st Author's Affiliation ()
()
32nd Author's Name / /
32nd Author's Affiliation ()
()
33rd Author's Name / /
33rd Author's Affiliation ()
()
34th Author's Name / /
34th Author's Affiliation ()
()
35th Author's Name / /
35th Author's Affiliation ()
()
36th Author's Name / /
36th Author's Affiliation ()
()
Speaker Author-1 
Date Time 2026-02-19 15:45:00 
Presentation Time 15 minutes 
Registration for ME 
Paper # MMS2026-16, ME2026-16, AIT2026-16, SIP2026-16 
Volume (vol) vol.50 
Number (no) no.5 
Page pp.70-74 
#Pages
Date of Issue 2026-02-12 (MMS, ME, AIT, SIP) 


[Return to Top Page]

[Return to ITE Web Page]


The Institute of Image Information and Television Engineers (ITE), Japan