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 2023-02-21 11:00
A note on text prompt tuning in cross-modal image retrieval for a specific database
Huaying Zhang, Rintaro Yanagi, Ren Togo, Takahiro Ogawa, Miki Haseyama (Hokkaido Univ.)
Abstract (in Japanese) (See Japanese page) 
(in English) With the development of storage devices and the Internet, the number of users creating personal image databases has increased. To effectively retrieve images from these databases, cross-modal image retrieval methods which allow users to retrieve images by entering a simple text query have been widely researched. Among these methods, several large cross-modal models pre-trained on huge amounts of image data have been proposed. However, to adapt these models to personal databases, it is required to store and update millions of parameters. This can be inefficient in terms of data storage in practice. To solve this problem, in this paper, we propose a novel image retrieval method that uses text prompt tuning to efficiently adapt the cross-modal model to a specific database without fine-tuning the model parameters. In the proposed method, we first construct a vector of several dimensions (prompt) and combine the prompt with the query vectorized to the same dimension. Then the combined vector and the candidate images are fed into the pre-trained cross-modal models. Finally, the prompt is optimized to achieve better retrieval accuracy. At the end of this paper, we have verified the effectiveness of the proposed method through experiments on the open dataset.
Keyword (in Japanese) (See Japanese page) 
(in English) Image retrieval / Cross-modal retrieval / Prompt learning / / / / /  
Reference Info. ITE Tech. Rep., vol. 47, no. 6, ME2023-23, pp. 11-15, Feb. 2023.
Paper # ME2023-23 
Date of Issue 2023-02-14 (MMS, ME, AIT) 
ISSN Print edition: ISSN 1342-6893    Online edition: ISSN 2424-1970
Download PDF

Conference Information
Conference Date 2023-02-21 - 2023-02-22 
Place (in Japanese) (See Japanese page) 
Place (in English) Hokkaido Univ. 
Topics (in Japanese) (See Japanese page) 
Topics (in English) Image Processing, etc. 
Paper Information
Registration To ME 
Conference Code 2023-02-MMS-ME-AIT-IE-ITS 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) A note on text prompt tuning in cross-modal image retrieval for a specific database 
Sub Title (in English)  
Keyword(1) Image retrieval  
Keyword(2) Cross-modal retrieval  
Keyword(3) Prompt learning  
1st Author's Name Huaying Zhang  
1st Author's Affiliation Hokkaido University (Hokkaido Univ.)
2nd Author's Name Rintaro Yanagi  
2nd Author's Affiliation Hokkaido University (Hokkaido Univ.)
3rd Author's Name Ren Togo  
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 ()
Speaker Author-1 
Date Time 2023-02-21 11:00:00 
Presentation Time 15 minutes 
Registration for ME 
Paper # MMS2023-3, ME2023-23, AIT2023-3 
Volume (vol) vol.47 
Number (no) no.6 
Page pp.11-15 
Date of Issue 2023-02-14 (MMS, ME, AIT) 

[Return to Top Page]

[Return to ITE Web Page]

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