Scipediacontent: Scipediacontent moved page Draft Content 563819791 to McLoughlin et al 2018a

2021-02-03T17:35:13Z

Scipediacontent moved page Draft Content 563819791 to McLoughlin et al 2018a

Scipediacontent: Created page with " == Abstract == There is a common observation that audio event classification is easier to deal with than detection. So far, this observation has been accepted as a fact and..."

2021-02-03T17:35:10Z

Created page with " == Abstract == There is a common observation that audio event classification is easier to deal with than detection. So far, this observation has been accepted as a fact and..."

New page

== Abstract ==

There is a common observation that audio event classification is easier to deal with than detection. So far, this observation has been accepted as a fact and we lack of a careful analysis. In this paper, we reason the rationale behind this fact and, more importantly, leverage them to benefit the audio event detection task. We present an improved detection pipeline in which a verification step is appended to augment a detection system. This step employs a high-quality event classifier to postprocess the benign event hypotheses outputted by the detection system and reject false alarms. To demonstrate the effectiveness of the proposed pipeline, we implement and pair up different event detectors based on the most common detection schemes and various event classifiers, ranging from the standard bag-of-words model to the state-of-the-art bank-of-regressors one. Experimental results on the ITC-Irst dataset show significant improvements to detection performance. More importantly, these improvements are consistent for all detector-classifier combinations.

Comment: Published version available at https://ieeexplore.ieee.org/document/8081709

Document type: Conference object

== Full document ==
<pdf>Media:Draft_Content_563819791-beopen5966-8393-document.pdf</pdf>

== Original document ==

The different versions of the original document can be found in:

* [http://arxiv.org/abs/1612.09089 http://arxiv.org/abs/1612.09089]

* [https://kar.kent.ac.uk/66057/1/WhatMakesAEDHarderThanClassification.pdf https://kar.kent.ac.uk/66057/1/WhatMakesAEDHarderThanClassification.pdf]

* [https://kar.kent.ac.uk/66057/31/phan2017d.pdf https://kar.kent.ac.uk/66057/31/phan2017d.pdf]

* [https://zenodo.org/record/1159698 https://zenodo.org/record/1159698] under the license http://creativecommons.org/licenses/by/4.0/legalcode

* [http://dx.doi.org/10.5281/zenodo.1159697 http://dx.doi.org/10.5281/zenodo.1159697],
: [https://zenodo.org/record/1159698 https://zenodo.org/record/1159698] under the license https://creativecommons.org/licenses/by/4.0

* [http://dx.doi.org/10.5281/zenodo.1159698 http://dx.doi.org/10.5281/zenodo.1159698],
: [https://zenodo.org/record/1159698 https://zenodo.org/record/1159698] under the license https://creativecommons.org/licenses/by/4.0

* [http://xplorestaging.ieee.org/ielx7/8067434/8081145/08081709.pdf?arnumber=8081709 http://xplorestaging.ieee.org/ielx7/8067434/8081145/08081709.pdf?arnumber=8081709],
: [http://dx.doi.org/10.23919/eusipco.2017.8081709 http://dx.doi.org/10.23919/eusipco.2017.8081709]

* [https://dblp.uni-trier.de/db/journals/corr/corr1612.html#PhanKMMMM16 https://dblp.uni-trier.de/db/journals/corr/corr1612.html#PhanKMMMM16],
: [https://arxiv.org/pdf/1612.09089.pdf https://arxiv.org/pdf/1612.09089.pdf],
: [https://arxiv.org/abs/1612.09089 https://arxiv.org/abs/1612.09089],
: [http://ui.adsabs.harvard.edu/abs/2016arXiv161209089P/abstract http://ui.adsabs.harvard.edu/abs/2016arXiv161209089P/abstract],
: [https://ieeexplore.ieee.org/document/8081709 https://ieeexplore.ieee.org/document/8081709],
: [https://za.arxiv.org/abs/1612.09089 https://za.arxiv.org/abs/1612.09089],
: [https://tw.arxiv.org/abs/1612.09089?context=cs https://tw.arxiv.org/abs/1612.09089?context=cs],
: [https://il.arxiv.org/abs/1612.09089 https://il.arxiv.org/abs/1612.09089],
: [https://jp.arxiv.org/abs/1612.09089 https://jp.arxiv.org/abs/1612.09089],
: [https://academic.microsoft.com/#/detail/2562770010 https://academic.microsoft.com/#/detail/2562770010]

DOIS: 10.5281/zenodo.1159698 10.5281/zenodo.1159697 10.23919/eusipco.2017.8081709

← Older revision	Revision as of 17:35, 3 February 2021
(No difference)

McLoughlin et al 2018a - Revision history

Scipediacontent: Scipediacontent moved page Draft Content 563819791 to McLoughlin et al 2018a

Scipediacontent: Created page with " == Abstract == There is a common observation that audio event classification is easier to deal with than detection. So far, this observation has been accepted as a fact and..."