Ensembling object detectors for image and video data analysis

Kateryna Chumachenko; Jenni Raitoharju; Alexandros Iosifidis; Moncef Gabbouj

doi:10.1109/ICASSP39728.2021.9414013

Ensembling object detectors for image and video data analysis

Kateryna Chumachenko, Jenni Raitoharju, Alexandros Iosifidis, Moncef Gabbouj

Tietotekniikka

Tutkimustuotos: Konferenssiartikkeli › Scientific › vertaisarvioitu

Abstrakti

In this paper, we propose a method for ensembling the outputs of multiple object detectors for improving detection performance and precision of bounding boxes on image data. We further extend it to video data by proposing a two-stage tracking-based scheme for detection refinement. The proposed method can be used as a standalone approach for improving object detection performance, or as a part of a framework for faster bounding box annotation in unseen datasets, assuming that the objects of interest are those present in some common public datasets.

Alkuperäiskieli	Englanti
Otsikko	ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Kustantaja	IEEE
Sivut	1515-1519
Sivumäärä	5
Vuosikerta	2021-June
ISBN (elektroninen)	978-1-7281-7605-5
DOI - pysyväislinkit	https://doi.org/10.1109/ICASSP39728.2021.9414013
Tila	Julkaistu - 2021
OKM-julkaisutyyppi	A4 Artikkeli konferenssijulkaisussa
Tapahtuma	IEEE International Conference on Acoustics, Speech and Signal Processing - Metro Toronto Convention Centre, Toronto, Kanada Kesto: 6 kesäk. 2021 → 11 kesäk. 2021 https://2021.ieeeicassp.org

Julkaisusarja

Nimi	Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing
ISSN (painettu)	1520-6149
ISSN (elektroninen)	2379-190X

Conference

Conference	IEEE International Conference on Acoustics, Speech and Signal Processing
Maa/Alue	Kanada
Kaupunki	Toronto
Ajanjakso	6/06/21 → 11/06/21
www-osoite	https://2021.ieeeicassp.org

Julkaisufoorumi-taso

Jufo-taso 1

!!ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Pääsy asiakirjaan

10.1109/ICASSP39728.2021.9414013

ICASSP_detection_emsemble
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Accepted author manuscript, 485 KB

https://urn.fi/URN:NBN:fi:tuni-202302102246

Siteeraa tätä

Chumachenko, K., Raitoharju, J., Iosifidis, A., & Gabbouj, M. (2021). Ensembling object detectors for image and video data analysis. teoksessa ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (Vuosikerta 2021-June, Sivut 1515-1519). (Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing). IEEE. https://doi.org/10.1109/ICASSP39728.2021.9414013

Chumachenko, Kateryna ; Raitoharju, Jenni ; Iosifidis, Alexandros et al. / Ensembling object detectors for image and video data analysis. ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Vuosikerta 2021-June IEEE, 2021. Sivut 1515-1519 (Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing).

@inproceedings{ee9fe58277d94efba66b43396a2a70fb,

title = "Ensembling object detectors for image and video data analysis",

abstract = "In this paper, we propose a method for ensembling the outputs of multiple object detectors for improving detection performance and precision of bounding boxes on image data. We further extend it to video data by proposing a two-stage tracking-based scheme for detection refinement. The proposed method can be used as a standalone approach for improving object detection performance, or as a part of a framework for faster bounding box annotation in unseen datasets, assuming that the objects of interest are those present in some common public datasets.",

keywords = "Bounding box annotation, Ensemble models, Object detection",

author = "Kateryna Chumachenko and Jenni Raitoharju and Alexandros Iosifidis and Moncef Gabbouj",

note = "Funding Information: This work was supported by Business Finland under the project 5G Vertical Integrated Industry for Massive Automation (5G-VIIMA). Publisher Copyright: {\textcopyright} 2021 IEEE jufoid=57409; IEEE International Conference on Acoustics, Speech and Signal Processing ; Conference date: 06-06-2021 Through 11-06-2021",

year = "2021",

doi = "10.1109/ICASSP39728.2021.9414013",

language = "English",

volume = "2021-June",

series = "Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing",

publisher = "IEEE",

pages = "1515--1519",

booktitle = "ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)",

url = "https://2021.ieeeicassp.org",

}

Chumachenko, K, Raitoharju, J, Iosifidis, A & Gabbouj, M 2021, Ensembling object detectors for image and video data analysis. julkaisussa ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Vuosikerta 2021-June, Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing, IEEE, Sivut 1515-1519, IEEE International Conference on Acoustics, Speech and Signal Processing, Toronto, Ontario, Kanada, 6/06/21. https://doi.org/10.1109/ICASSP39728.2021.9414013

Ensembling object detectors for image and video data analysis. / Chumachenko, Kateryna; Raitoharju, Jenni; Iosifidis, Alexandros et al.

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Vuosikerta 2021-June IEEE, 2021. s. 1515-1519 (Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing).

Tutkimustuotos: Konferenssiartikkeli › Scientific › vertaisarvioitu

TY - GEN

T1 - Ensembling object detectors for image and video data analysis

AU - Chumachenko, Kateryna

AU - Raitoharju, Jenni

AU - Iosifidis, Alexandros

AU - Gabbouj, Moncef

PY - 2021

Y1 - 2021

N2 - In this paper, we propose a method for ensembling the outputs of multiple object detectors for improving detection performance and precision of bounding boxes on image data. We further extend it to video data by proposing a two-stage tracking-based scheme for detection refinement. The proposed method can be used as a standalone approach for improving object detection performance, or as a part of a framework for faster bounding box annotation in unseen datasets, assuming that the objects of interest are those present in some common public datasets.

AB - In this paper, we propose a method for ensembling the outputs of multiple object detectors for improving detection performance and precision of bounding boxes on image data. We further extend it to video data by proposing a two-stage tracking-based scheme for detection refinement. The proposed method can be used as a standalone approach for improving object detection performance, or as a part of a framework for faster bounding box annotation in unseen datasets, assuming that the objects of interest are those present in some common public datasets.

KW - Bounding box annotation

KW - Ensemble models

KW - Object detection

U2 - 10.1109/ICASSP39728.2021.9414013

DO - 10.1109/ICASSP39728.2021.9414013

M3 - Conference contribution

AN - SCOPUS:85115180582

VL - 2021-June

T3 - Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing

SP - 1515

EP - 1519

BT - ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

PB - IEEE

T2 - IEEE International Conference on Acoustics, Speech and Signal Processing

Y2 - 6 June 2021 through 11 June 2021

ER -

Chumachenko K, Raitoharju J, Iosifidis A, Gabbouj M. Ensembling object detectors for image and video data analysis. julkaisussa ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Vuosikerta 2021-June. IEEE. 2021. s. 1515-1519. (Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing). doi: 10.1109/ICASSP39728.2021.9414013

Ensembling object detectors for image and video data analysis

Abstrakti

Julkaisusarja

Conference

Julkaisufoorumi-taso

!!ASJC Scopus subject areas

Pääsy asiakirjaan

Sormenjälki

Siteeraa tätä