Comparative analysis of relevance feedback techniques for image retrieval

Lucia Vadicamo; Francesca Scotti; Alan Dearle; Richard Connor

doi:10.1007/978-981-96-2054-8_16

Comparative analysis of relevance feedback techniques for image retrieval

Lucia Vadicamo^*, Francesca Scotti, Alan Dearle, Richard Connor

^*Corresponding author for this work

School of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Relevance feedback mechanisms have garnered significant attention in content-based image and video retrieval thanks to their effectiveness in refining search results to better meet user information needs. This paper provides a comprehensive comparative analysis of four techniques: Rocchio, PicHunter, Polyadic Query, and linear Support Vector Machines, representing diverse strategies encompassing query vector modification, relevance probability estimation, adaptive similarity metrics, and classifier learning. We conducted experiments within an interactive image retrieval system, with varying amounts of user feedback: full feedback, limited positive feedback, and mixed feedback. In particular, we introduce novel enhanced versions of PicHunter and Polyadic search incorporating negative feedback. Our findings highlight the benefits of integrating both positive and negative examples, demonstrating significant performance improvements. Overall, SVM and our improved PicHunter outperformed the other approaches for ad-hoc search, especially in cases in which the feedback process is iterated several times.

Original language	English
Title of host publication	MultiMedia modeling
Subtitle of host publication	31st international conference on multimedia modeling, MMM 2025, Proceedings
Editors	Ichiro Ide, Ioannis Kompatsiaris, Changsheng Xu, Keiji Yanai, Wei-Ta Chu, Naoko Nitta, Michael Riegler, Toshihiko Yamasaki
Place of Publication	Singapore
Publisher	Springer Singapore
Pages	206-219
Number of pages	14
ISBN (Electronic)	9789819620548
ISBN (Print)	9789819620531
DOIs	https://doi.org/10.1007/978-981-96-2054-8_16
Publication status	Published - 3 Jan 2025
Event	31st International Conference on Multimedia Modeling, MMM 2025 - Nara, Japan Duration: 8 Jan 2025 → 10 Jan 2025

Publication series

Name	Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and Lecture notes in bioinformatics)
Publisher	Springer Nature
Volume	15520 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	31st International Conference on Multimedia Modeling, MMM 2025
Country/Territory	Japan
City	Nara
Period	8/01/25 → 10/01/25

Keywords

Content-based image retrieval
PicHunter
Polyadic query
Relevance feedback
Rocchio
SVM

Access to Document

10.1007/978-981-96-2054-8_16

Handle.net

Persistent link

Cite this

Vadicamo, L., Scotti, F., Dearle, A., & Connor, R. (2025). Comparative analysis of relevance feedback techniques for image retrieval. In I. Ide, I. Kompatsiaris, C. Xu, K. Yanai, W.-T. Chu, N. Nitta, M. Riegler, & T. Yamasaki (Eds.), MultiMedia modeling: 31st international conference on multimedia modeling, MMM 2025, Proceedings (pp. 206-219). (Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and Lecture notes in bioinformatics); Vol. 15520 LNCS). Springer Singapore. https://doi.org/10.1007/978-981-96-2054-8_16

Vadicamo, Lucia ; Scotti, Francesca ; Dearle, Alan et al. / Comparative analysis of relevance feedback techniques for image retrieval. MultiMedia modeling: 31st international conference on multimedia modeling, MMM 2025, Proceedings. editor / Ichiro Ide ; Ioannis Kompatsiaris ; Changsheng Xu ; Keiji Yanai ; Wei-Ta Chu ; Naoko Nitta ; Michael Riegler ; Toshihiko Yamasaki. Singapore : Springer Singapore, 2025. pp. 206-219 (Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and Lecture notes in bioinformatics)).

@inproceedings{27eb33568af945c4b254fdeb94bd0d67,

title = "Comparative analysis of relevance feedback techniques for image retrieval",

abstract = "Relevance feedback mechanisms have garnered significant attention in content-based image and video retrieval thanks to their effectiveness in refining search results to better meet user information needs. This paper provides a comprehensive comparative analysis of four techniques: Rocchio, PicHunter, Polyadic Query, and linear Support Vector Machines, representing diverse strategies encompassing query vector modification, relevance probability estimation, adaptive similarity metrics, and classifier learning. We conducted experiments within an interactive image retrieval system, with varying amounts of user feedback: full feedback, limited positive feedback, and mixed feedback. In particular, we introduce novel enhanced versions of PicHunter and Polyadic search incorporating negative feedback. Our findings highlight the benefits of integrating both positive and negative examples, demonstrating significant performance improvements. Overall, SVM and our improved PicHunter outperformed the other approaches for ad-hoc search, especially in cases in which the feedback process is iterated several times.",

keywords = "Content-based image retrieval, PicHunter, Polyadic query, Relevance feedback, Rocchio, SVM",

author = "Lucia Vadicamo and Francesca Scotti and Alan Dearle and Richard Connor",

year = "2025",

month = jan,

day = "3",

doi = "10.1007/978-981-96-2054-8\_16",

language = "English",

isbn = "9789819620531",

series = "Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and Lecture notes in bioinformatics)",

publisher = "Springer Singapore",

pages = "206--219",

editor = "Ichiro Ide and Ioannis Kompatsiaris and Changsheng Xu and Keiji Yanai and Wei-Ta Chu and Naoko Nitta and Michael Riegler and Toshihiko Yamasaki",

booktitle = "MultiMedia modeling",

note = "31st International Conference on Multimedia Modeling, MMM 2025 ; Conference date: 08-01-2025 Through 10-01-2025",

}

Vadicamo, L, Scotti, F, Dearle, A & Connor, R 2025, Comparative analysis of relevance feedback techniques for image retrieval. in I Ide, I Kompatsiaris, C Xu, K Yanai, W-T Chu, N Nitta, M Riegler & T Yamasaki (eds), MultiMedia modeling: 31st international conference on multimedia modeling, MMM 2025, Proceedings. Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and Lecture notes in bioinformatics), vol. 15520 LNCS, Springer Singapore, Singapore, pp. 206-219, 31st International Conference on Multimedia Modeling, MMM 2025, Nara, Japan, 8/01/25. https://doi.org/10.1007/978-981-96-2054-8_16

Comparative analysis of relevance feedback techniques for image retrieval. / Vadicamo, Lucia; Scotti, Francesca; Dearle, Alan et al.
MultiMedia modeling: 31st international conference on multimedia modeling, MMM 2025, Proceedings. ed. / Ichiro Ide; Ioannis Kompatsiaris; Changsheng Xu; Keiji Yanai; Wei-Ta Chu; Naoko Nitta; Michael Riegler; Toshihiko Yamasaki. Singapore: Springer Singapore, 2025. p. 206-219 (Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and Lecture notes in bioinformatics); Vol. 15520 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Comparative analysis of relevance feedback techniques for image retrieval

AU - Vadicamo, Lucia

AU - Scotti, Francesca

AU - Dearle, Alan

AU - Connor, Richard

PY - 2025/1/3

Y1 - 2025/1/3

N2 - Relevance feedback mechanisms have garnered significant attention in content-based image and video retrieval thanks to their effectiveness in refining search results to better meet user information needs. This paper provides a comprehensive comparative analysis of four techniques: Rocchio, PicHunter, Polyadic Query, and linear Support Vector Machines, representing diverse strategies encompassing query vector modification, relevance probability estimation, adaptive similarity metrics, and classifier learning. We conducted experiments within an interactive image retrieval system, with varying amounts of user feedback: full feedback, limited positive feedback, and mixed feedback. In particular, we introduce novel enhanced versions of PicHunter and Polyadic search incorporating negative feedback. Our findings highlight the benefits of integrating both positive and negative examples, demonstrating significant performance improvements. Overall, SVM and our improved PicHunter outperformed the other approaches for ad-hoc search, especially in cases in which the feedback process is iterated several times.

AB - Relevance feedback mechanisms have garnered significant attention in content-based image and video retrieval thanks to their effectiveness in refining search results to better meet user information needs. This paper provides a comprehensive comparative analysis of four techniques: Rocchio, PicHunter, Polyadic Query, and linear Support Vector Machines, representing diverse strategies encompassing query vector modification, relevance probability estimation, adaptive similarity metrics, and classifier learning. We conducted experiments within an interactive image retrieval system, with varying amounts of user feedback: full feedback, limited positive feedback, and mixed feedback. In particular, we introduce novel enhanced versions of PicHunter and Polyadic search incorporating negative feedback. Our findings highlight the benefits of integrating both positive and negative examples, demonstrating significant performance improvements. Overall, SVM and our improved PicHunter outperformed the other approaches for ad-hoc search, especially in cases in which the feedback process is iterated several times.

KW - Content-based image retrieval

KW - PicHunter

KW - Polyadic query

KW - Relevance feedback

KW - Rocchio

KW - SVM

UR - https://www.scopus.com/pages/publications/85216124971

U2 - 10.1007/978-981-96-2054-8_16

DO - 10.1007/978-981-96-2054-8_16

M3 - Conference contribution

AN - SCOPUS:85216124971

SN - 9789819620531

T3 - Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and Lecture notes in bioinformatics)

SP - 206

EP - 219

BT - MultiMedia modeling

A2 - Ide, Ichiro

A2 - Kompatsiaris, Ioannis

A2 - Xu, Changsheng

A2 - Yanai, Keiji

A2 - Chu, Wei-Ta

A2 - Nitta, Naoko

A2 - Riegler, Michael

A2 - Yamasaki, Toshihiko

PB - Springer Singapore

CY - Singapore

T2 - 31st International Conference on Multimedia Modeling, MMM 2025

Y2 - 8 January 2025 through 10 January 2025

ER -

Vadicamo L, Scotti F, Dearle A , Connor R. Comparative analysis of relevance feedback techniques for image retrieval. In Ide I, Kompatsiaris I, Xu C, Yanai K, Chu WT, Nitta N, Riegler M, Yamasaki T, editors, MultiMedia modeling: 31st international conference on multimedia modeling, MMM 2025, Proceedings. Singapore: Springer Singapore. 2025. p. 206-219. (Lecture notes in computer science (including subseries Lecture notes in artificial intelligence and Lecture notes in bioinformatics)). doi: 10.1007/978-981-96-2054-8_16

Comparative analysis of relevance feedback techniques for image retrieval

Abstract

Publication series

Conference

Keywords

Access to Document

Handle.net

Other files and links

Fingerprint

ADR UK Programme: University of Edinburgh 2022-2026 ADR UK Programme

Cite this

Comparative analysis of relevance feedback techniques for image retrieval

Abstract

Publication series

Conference

Keywords

Access to Document

Handle.net

Other files and links

Fingerprint

Projects

ADR UK Programme: University of Edinburgh 2022-2026 ADR UK Programme

Cite this