A Local Feature Aggregation Method for Music Retrieval
Summary :
The song-level feature summarization is an essential building block for browsing, retrieval, and indexing of digital music. This paper proposes a local pooling method to aggregate the feature vectors of a song over the universal background model. Two types of local activation patterns of feature vectors are derived; one representation is derived in the form of histogram, and the other is given by a binary vector. Experiments over three publicly-available music datasets show that the proposed local aggregation of the auditory features is promising for music-similarity computation.
- Publication
- IEICE TRANSACTIONS on Information Vol.E101-D No.1 pp.64-67
- Publication Date
- 2018/01/01
- Publicized
- 2017/10/16
- Online ISSN
- 1745-1361
- DOI
- 10.1587/transinf.2017MUL0001
- Type of Manuscript
- Special Section LETTER (Special Section on Enriched Multimedia — Potential and Possibility of Multimedia Contents for the Future —)
- Category
Authors
Jin S. SEO
Gangneung-Wonju National University
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Jin S. SEO, "A Local Feature Aggregation Method for Music Retrieval" in IEICE TRANSACTIONS on Information,
vol. E101-D, no. 1, pp. 64-67, January 2018, doi: 10.1587/transinf.2017MUL0001.
Abstract: The song-level feature summarization is an essential building block for browsing, retrieval, and indexing of digital music. This paper proposes a local pooling method to aggregate the feature vectors of a song over the universal background model. Two types of local activation patterns of feature vectors are derived; one representation is derived in the form of histogram, and the other is given by a binary vector. Experiments over three publicly-available music datasets show that the proposed local aggregation of the auditory features is promising for music-similarity computation.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2017MUL0001/_p
Copy
@ARTICLE{e101-d_1_64,
author={Jin S. SEO, },
journal={IEICE TRANSACTIONS on Information},
title={A Local Feature Aggregation Method for Music Retrieval},
year={2018},
volume={E101-D},
number={1},
pages={64-67},
abstract={The song-level feature summarization is an essential building block for browsing, retrieval, and indexing of digital music. This paper proposes a local pooling method to aggregate the feature vectors of a song over the universal background model. Two types of local activation patterns of feature vectors are derived; one representation is derived in the form of histogram, and the other is given by a binary vector. Experiments over three publicly-available music datasets show that the proposed local aggregation of the auditory features is promising for music-similarity computation.},
keywords={},
doi={10.1587/transinf.2017MUL0001},
ISSN={1745-1361},
month={January},}
Copy
TY - JOUR
TI - A Local Feature Aggregation Method for Music Retrieval
T2 - IEICE TRANSACTIONS on Information
SP - 64
EP - 67
AU - Jin S. SEO
PY - 2018
DO - 10.1587/transinf.2017MUL0001
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E101-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2018
AB - The song-level feature summarization is an essential building block for browsing, retrieval, and indexing of digital music. This paper proposes a local pooling method to aggregate the feature vectors of a song over the universal background model. Two types of local activation patterns of feature vectors are derived; one representation is derived in the form of histogram, and the other is given by a binary vector. Experiments over three publicly-available music datasets show that the proposed local aggregation of the auditory features is promising for music-similarity computation.
ER -
English
