The amount of digital material in video lecture archives is growing rapidly, causing the search&retrieval process to be time-consuming and almost impractical. Indeed, after the search, students receive a list of videos and often must use VCR-like functions to find the specific piece of video that covers the searched topic. Therefore, a more efficient method for video retrieval in digital video lecture archives is needed. In this paper, we propose VLB (Video Lecture Browsing), a system designed to facilitate both the retrieval of video lectures within video archives and the finding of the most appropriate segment of a video lecture that covers a searched topic by automatically producing a general picture of the contents of a video lecture. To achieve these goals, the system introduces the idea of timed tag-clouds, which are produced with a combination of aural and visual analysis. Results of a MOS evaluation show that users highly appreciate the timed tag-clouds approach and a comparison study against other popular approaches shows that 93 % of users prefer to use VLB to handle video lectures.

On introducing timed tag-clouds in video lectures indexing / Furini, Marco. - In: MULTIMEDIA TOOLS AND APPLICATIONS. - ISSN 1380-7501. - 77:1(2018), pp. 967-984. [10.1007/s11042-016-4282-5]

On introducing timed tag-clouds in video lectures indexing

FURINI, Marco
2018

Abstract

The amount of digital material in video lecture archives is growing rapidly, causing the search&retrieval process to be time-consuming and almost impractical. Indeed, after the search, students receive a list of videos and often must use VCR-like functions to find the specific piece of video that covers the searched topic. Therefore, a more efficient method for video retrieval in digital video lecture archives is needed. In this paper, we propose VLB (Video Lecture Browsing), a system designed to facilitate both the retrieval of video lectures within video archives and the finding of the most appropriate segment of a video lecture that covers a searched topic by automatically producing a general picture of the contents of a video lecture. To achieve these goals, the system introduces the idea of timed tag-clouds, which are produced with a combination of aural and visual analysis. Results of a MOS evaluation show that users highly appreciate the timed tag-clouds approach and a comparison study against other popular approaches shows that 93 % of users prefer to use VLB to handle video lectures.
2018
6-gen-2017
77
1
967
984
On introducing timed tag-clouds in video lectures indexing / Furini, Marco. - In: MULTIMEDIA TOOLS AND APPLICATIONS. - ISSN 1380-7501. - 77:1(2018), pp. 967-984. [10.1007/s11042-016-4282-5]
Furini, Marco
File in questo prodotto:
File Dimensione Formato  
indexing-tag-clouds.pdf

Accesso riservato

Tipologia: Versione dell'autore revisionata e accettata per la pubblicazione
Dimensione 762.32 kB
Formato Adobe PDF
762.32 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

Licenza Creative Commons
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11380/1133574
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 17
  • ???jsp.display-item.citation.isi??? 12
social impact