Random forest classification based acoustic event detection

Xianjun Xia; Roberto Togneri; Ferdous Sohel; David Huang

doi:10.1109/ICME.2017.8019452

Random forest classification based acoustic event detection

Xianjun Xia, Roberto Togneri, Ferdous Sohel, David Huang

Research output: Chapter in Book/Conference paper › Conference paper › peer-review

19 Citations (Scopus)

Abstract

This paper deals with the acoustic event detection (AED) to improve the detection accuracy of acoustic events. Acoustic event detection task is performed by a regression via classification (RvC) based approach along with the random forest technique. A discretization process is used to convert the continuous frame positions within acoustic events into event duration class labels. Outputs of the category-specific random forest classifiers are then reversed back to the event boundary information. Evaluations on the UPC-TALP database which consists of highly variable acoustic events demonstrate the efficiency of the proposed approaches with improvements in detection error rate compared to the best baseline system.

Original language	English
Title of host publication	Proceedings of the IEEE International Conference on Multimedia and Expo 2017
Editors	Jörn Ostermann, Kenneth K.M. Lam
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	163-168
Number of pages	6
ISBN (Print)	9781509060672
DOIs	https://doi.org/10.1109/ICME.2017.8019452
Publication status	Published - 31 Aug 2017
Event	2017 IEEE International Conference on Multimedia and Expo, ICME 2017 - Hong Kong, Hong Kong Duration: 10 Jul 2017 → 14 Jul 2017

Conference

Conference	2017 IEEE International Conference on Multimedia and Expo, ICME 2017
Country/Territory	Hong Kong
City	Hong Kong
Period	10/07/17 → 14/07/17

Access to Document

10.1109/ICME.2017.8019452

Cite this

@inproceedings{ec4e5793e4f34e909df3af0e7838b24c,

title = "Random forest classification based acoustic event detection",

abstract = "This paper deals with the acoustic event detection (AED) to improve the detection accuracy of acoustic events. Acoustic event detection task is performed by a regression via classification (RvC) based approach along with the random forest technique. A discretization process is used to convert the continuous frame positions within acoustic events into event duration class labels. Outputs of the category-specific random forest classifiers are then reversed back to the event boundary information. Evaluations on the UPC-TALP database which consists of highly variable acoustic events demonstrate the efficiency of the proposed approaches with improvements in detection error rate compared to the best baseline system.",

keywords = "Acoustic event detection, Discretization, Random forest, Regression via classification (RvC)",

author = "Xianjun Xia and Roberto Togneri and Ferdous Sohel and David Huang",

year = "2017",

month = aug,

day = "31",

doi = "10.1109/ICME.2017.8019452",

language = "English",

isbn = "9781509060672",

pages = "163--168",

editor = "Ostermann, {J{\"o}rn } and Lam, {Kenneth K.M. }",

booktitle = "Proceedings of the IEEE International Conference on Multimedia and Expo 2017",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States",

note = "2017 IEEE International Conference on Multimedia and Expo, ICME 2017 ; Conference date: 10-07-2017 Through 14-07-2017",

}

Xia, X, Togneri, R, Sohel, F & Huang, D 2017, Random forest classification based acoustic event detection. in J Ostermann & KKM Lam (eds), Proceedings of the IEEE International Conference on Multimedia and Expo 2017., 8019452, IEEE, Institute of Electrical and Electronics Engineers, pp. 163-168, 2017 IEEE International Conference on Multimedia and Expo, ICME 2017, Hong Kong, Hong Kong, 10/07/17. https://doi.org/10.1109/ICME.2017.8019452

Random forest classification based acoustic event detection. / Xia, Xianjun; Togneri, Roberto; Sohel, Ferdous et al.
Proceedings of the IEEE International Conference on Multimedia and Expo 2017. ed. / Jörn Ostermann; Kenneth K.M. Lam. IEEE, Institute of Electrical and Electronics Engineers, 2017. p. 163-168 8019452.

Research output: Chapter in Book/Conference paper › Conference paper › peer-review

TY - GEN

T1 - Random forest classification based acoustic event detection

AU - Xia, Xianjun

AU - Togneri, Roberto

AU - Sohel, Ferdous

AU - Huang, David

PY - 2017/8/31

Y1 - 2017/8/31

N2 - This paper deals with the acoustic event detection (AED) to improve the detection accuracy of acoustic events. Acoustic event detection task is performed by a regression via classification (RvC) based approach along with the random forest technique. A discretization process is used to convert the continuous frame positions within acoustic events into event duration class labels. Outputs of the category-specific random forest classifiers are then reversed back to the event boundary information. Evaluations on the UPC-TALP database which consists of highly variable acoustic events demonstrate the efficiency of the proposed approaches with improvements in detection error rate compared to the best baseline system.

AB - This paper deals with the acoustic event detection (AED) to improve the detection accuracy of acoustic events. Acoustic event detection task is performed by a regression via classification (RvC) based approach along with the random forest technique. A discretization process is used to convert the continuous frame positions within acoustic events into event duration class labels. Outputs of the category-specific random forest classifiers are then reversed back to the event boundary information. Evaluations on the UPC-TALP database which consists of highly variable acoustic events demonstrate the efficiency of the proposed approaches with improvements in detection error rate compared to the best baseline system.

KW - Acoustic event detection

KW - Discretization

KW - Random forest

KW - Regression via classification (RvC)

UR - http://www.scopus.com/inward/record.url?scp=85030261128&partnerID=8YFLogxK

U2 - 10.1109/ICME.2017.8019452

DO - 10.1109/ICME.2017.8019452

M3 - Conference paper

AN - SCOPUS:85030261128

SN - 9781509060672

SP - 163

EP - 168

BT - Proceedings of the IEEE International Conference on Multimedia and Expo 2017

A2 - Ostermann, Jörn

A2 - Lam, Kenneth K.M.

PB - IEEE, Institute of Electrical and Electronics Engineers

T2 - 2017 IEEE International Conference on Multimedia and Expo, ICME 2017

Y2 - 10 July 2017 through 14 July 2017

ER -

Random forest classification based acoustic event detection

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this