DocumentCode :
2948384
Title :
SOLAR: sound object localization and retrieval in complex audio environments
Author :
Hoiem, Derek ; Ke, Yan ; Sukthankar, Rahul
Author_Institution :
Sch. of Comput. Sci., Carnegie Mellon Univ., Pittsburgh, PA, USA
Volume :
5
fYear :
2005
fDate :
18-23 March 2005
Abstract :
The ability to identify sounds in complex audio environments is highly useful for multimedia retrieval, security, and many mobile robotic applications, but very little work has been done in this area. We present the SOLAR system, a system capable of finding sound objects, such as dog barks or car horns, in complex audio data extracted from movies. SOLAR avoids the need for segmentation by scanning over the audio data in fixed increments and classifying each short audio window separately. SOLAR employs boosted decision tree classifiers to select suitable features for modeling each sound object and to discriminate between the object of interest and all other sounds. We demonstrate the effectiveness of our approach with experiments on thirteen sound object classes trained using only tens of positive examples and tested on hours of audio data extracted from popular movies.
Keywords :
audio signal processing; decision trees; signal classification; audio signal classification; boosted decision tree classifiers; complex audio environments; mobile robotic applications; multimedia retrieval; security; sound object localization and retrieval; sound object retrieval; Background noise; Computer science; Computer security; Data mining; Data security; Gunshot detection systems; Mobile robots; Motion pictures; Music; Object detection;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2005. Proceedings. (ICASSP '05). IEEE International Conference on
ISSN :
1520-6149
Print_ISBN :
0-7803-8874-7
Type :
conf
DOI :
10.1109/ICASSP.2005.1416332
Filename :
1416332
Link To Document :
بازگشت