Title :
Query sound-by-example video retrieval framework
Author :
Feki, Issam ; Ben Ammar, Anis ; Alimi, Adel M.
Author_Institution :
REGIM-Lab.: Res. Groups in Intell. Machines, Univ. of Sfax, Sfax, Tunisia
Abstract :
In this paper, query sound-by-example video retrieval framework based on audio concepts is presented. First, audio stream extracted from movies in the database is set into orientation clusters using an unsupervised segmentation technique. Audio signals admit a new proposed particular pretreatment process to distinguish audio concepts. This is used for indexing the video data. Second, the query asked by the user, in sound signal form, is treated. Finally, a specific retrieval function is used to obtain video shot containing the sound of query. Objective evaluation reached 89% retrieval performance.
Keywords :
audio signal processing; content-based retrieval; pattern clustering; source separation; video retrieval; video signal processing; audio concept; audio signals; audio stream extraction; movies; orientation clusters; query sound-by-example video retrieval framework; retrieval function; sound signal; unsupervised segmentation technique; video data indexing; video shot; Accuracy; Hidden Markov models; Indexing; Music; Silicon; Speech; Support vector machines; audio concept; query sound; sound indexing; video retrieval;
Conference_Titel :
Hybrid Intelligent Systems (HIS), 2014 14th International Conference on
Print_ISBN :
978-1-4799-7632-4
DOI :
10.1109/HIS.2014.7086165
