Exquisitor at the Video Browser Showdown 2025: Unifying Conversational Search and User Relevance Feedback

Ujjwal Sharma, Omar Shahbaz Khan, Stevan Rudinac, Björn Þór Jónsson

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Exquisitor is a multimedia retrieval system designed to combine conversational search and user relevance feedback to enhance interactive video search. In this work, we present recent improvements to Exquisitor, focusing on integrating these two search strategies into a unified, user-friendly interface. In this system edition set to participate at VBS 2025, we propose improvements for fusing results from the conversational search and user relevance feedback subsystems into a single search pane, an ability to launch multiple tabbed search instances and several improvements to the user interface to significantly improve user experience and search efficiency, particularly for novice users.

Original languageEnglish
Title of host publicationMultiMedia Modeling - 31st International Conference on Multimedia Modeling, MMM 2025, Proceedings
EditorsIchiro Ide, Ioannis Kompatsiaris, Changsheng Xu, Keiji Yanai, Wei-Ta Chu, Naoko Nitta, Michael Riegler, Toshihiko Yamasaki
PublisherSpringer Science and Business Media Deutschland GmbH
Pages264-271
Number of pages8
ISBN (Print)9789819620739
DOIs
Publication statusPublished - 2025
Event31st International Conference on Multimedia Modeling, MMM 2025 - Nara, Japan
Duration: 8 Jan 202510 Jan 2025

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume15524 LNCS

Conference

Conference31st International Conference on Multimedia Modeling, MMM 2025
Country/TerritoryJapan
CityNara
Period8/01/2510/01/25

Bibliographical note

Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

Other keywords

  • Interactive learning
  • Multimedia retrieval
  • Multimodal conversational search
  • Multimodal representation learning
  • Video browsing

Fingerprint

Dive into the research topics of 'Exquisitor at the Video Browser Showdown 2025: Unifying Conversational Search and User Relevance Feedback'. Together they form a unique fingerprint.

Cite this