KU-ISPL TRECVID 2015 multimedia event detection system

Seongjae Lee; Minseok Keum; Cheoljong Yang; Jeongmin Bae; Han Wang; Taeyup Song; Dubok Park; Daehun Kim; Jaeyong Ju; Hanseok Ko

KU-ISPL TRECVID 2015 multimedia event detection system

Seongjae Lee, Minseok Keum, Cheoljong Yang, Jeongmin Bae, Han Wang, Taeyup Song, Dubok Park, Daehun Kim, Jaeyong Ju, Hanseok Ko

School of Electrical Engineering

Research output: Contribution to conference › Paper › peer-review

Abstract

This paper describes the KU-ISPL Multimedia Event Detection (MED) system employed for competing in the TRECVID 2015 hosted by NIST. The proposed system utilizes diverse audio/visual information source components which consist of a combination of low-level and semantic-level features and adopts an optimized information fusion technique for accurate and robust video event detection. In particular, the local descriptors of the system are composed of the heterogeneous features extracted from the audio/visual information sources of video contents. The fusion process combines the information source at either low-level or semantic-level so that the individual detection score for video events can be judicially appraised in terms of meaningful representation and significance. The results from self-test and the official evaluation have indicated that the proposed system outperforms the previous version.

Original language	English
Publication status	Published - 2020
Event	2015 TREC Video Retrieval Evaluation, TRECVID 2015 - Gaithersburg, United States Duration: 2015 Nov 16 → 2015 Nov 18

Conference

Conference	2015 TREC Video Retrieval Evaluation, TRECVID 2015
Country/Territory	United States
City	Gaithersburg
Period	15/11/16 → 15/11/18

ASJC Scopus subject areas

Information Systems
Signal Processing
Electrical and Electronic Engineering

Cite this

@conference{a0b8c328177f4a448f80b36f7b44a94e,

title = "KU-ISPL TRECVID 2015 multimedia event detection system",

abstract = "This paper describes the KU-ISPL Multimedia Event Detection (MED) system employed for competing in the TRECVID 2015 hosted by NIST. The proposed system utilizes diverse audio/visual information source components which consist of a combination of low-level and semantic-level features and adopts an optimized information fusion technique for accurate and robust video event detection. In particular, the local descriptors of the system are composed of the heterogeneous features extracted from the audio/visual information sources of video contents. The fusion process combines the information source at either low-level or semantic-level so that the individual detection score for video events can be judicially appraised in terms of meaningful representation and significance. The results from self-test and the official evaluation have indicated that the proposed system outperforms the previous version.",

author = "Seongjae Lee and Minseok Keum and Cheoljong Yang and Jeongmin Bae and Han Wang and Taeyup Song and Dubok Park and Daehun Kim and Jaeyong Ju and Hanseok Ko",

note = "Publisher Copyright: {\textcopyright} 2015 TRECVID.; 2015 TREC Video Retrieval Evaluation, TRECVID 2015 ; Conference date: 16-11-2015 Through 18-11-2015",

year = "2020",

language = "English",

}

TY - CONF

T1 - KU-ISPL TRECVID 2015 multimedia event detection system

AU - Lee, Seongjae

AU - Keum, Minseok

AU - Yang, Cheoljong

AU - Bae, Jeongmin

AU - Wang, Han

AU - Song, Taeyup

AU - Park, Dubok

AU - Kim, Daehun

AU - Ju, Jaeyong

AU - Ko, Hanseok

PY - 2020

Y1 - 2020

N2 - This paper describes the KU-ISPL Multimedia Event Detection (MED) system employed for competing in the TRECVID 2015 hosted by NIST. The proposed system utilizes diverse audio/visual information source components which consist of a combination of low-level and semantic-level features and adopts an optimized information fusion technique for accurate and robust video event detection. In particular, the local descriptors of the system are composed of the heterogeneous features extracted from the audio/visual information sources of video contents. The fusion process combines the information source at either low-level or semantic-level so that the individual detection score for video events can be judicially appraised in terms of meaningful representation and significance. The results from self-test and the official evaluation have indicated that the proposed system outperforms the previous version.

AB - This paper describes the KU-ISPL Multimedia Event Detection (MED) system employed for competing in the TRECVID 2015 hosted by NIST. The proposed system utilizes diverse audio/visual information source components which consist of a combination of low-level and semantic-level features and adopts an optimized information fusion technique for accurate and robust video event detection. In particular, the local descriptors of the system are composed of the heterogeneous features extracted from the audio/visual information sources of video contents. The fusion process combines the information source at either low-level or semantic-level so that the individual detection score for video events can be judicially appraised in terms of meaningful representation and significance. The results from self-test and the official evaluation have indicated that the proposed system outperforms the previous version.

UR - http://www.scopus.com/inward/record.url?scp=85084730254&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85084730254

T2 - 2015 TREC Video Retrieval Evaluation, TRECVID 2015

Y2 - 16 November 2015 through 18 November 2015

ER -

KU-ISPL TRECVID 2015 multimedia event detection system

Abstract

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this