Outlier-Robust Neural Aggregation Network for Video Face Identification

Stefan Hormann; Martin Knoche; Maryam Babaee; Okan Kopuklu; Gerhard Rigoll

doi:10.1109/ICIP.2019.8803028

Outlier-Robust Neural Aggregation Network for Video Face Identification

Stefan Hormann, Martin Knoche, Maryam Babaee, Okan Kopuklu, Gerhard Rigoll

Chair of Human-Machine Communication

Technical University of Munich

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Scopus citations

Abstract

Current approaches for video face recognition rely on image sets containing faces of exclusively one identity. However, as image sets are created by unsupervised methods, it is necessary to consider outlier-afflicted sets for real-life applications. In this paper, we propose an Outlier-Robust Neural Aggregation Network (ORNAN). First, we embed each image into a feature space using a Convolutional Neural Network (CNN). With the help of two cascaded attention blocks, we predict outliers within the image set. By integrating this knowledge into our aggregation network, we adaptively aggregate all feature vectors to form a single feature, mitigating the influence of outliers and noisy features. We show that our network is robust against outliers using outlier-afflicted IJB-B and IJB-C benchmarks while maintaining similar performance without outliers.

Original language	English
Title of host publication	2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings
Publisher	IEEE Computer Society
Pages	1675-1679
Number of pages	5
ISBN (Electronic)	9781538662496
DOIs	https://doi.org/10.1109/ICIP.2019.8803028
State	Published - Sep 2019
Event	26th IEEE International Conference on Image Processing, ICIP 2019 - Taipei, Taiwan, Province of China Duration: 22 Sep 2019 → 25 Sep 2019

Publication series

Name	Proceedings - International Conference on Image Processing, ICIP
Volume	2019-September
ISSN (Print)	1522-4880

Conference

Conference	26th IEEE International Conference on Image Processing, ICIP 2019
Country/Territory	Taiwan, Province of China
City	Taipei
Period	22/09/19 → 25/09/19

Keywords

biometrics
face identification
feature aggregation
video face recognition

Access to Document

10.1109/ICIP.2019.8803028

Cite this

Hormann, S., Knoche, M., Babaee, M., Kopuklu, O., & Rigoll, G. (2019). Outlier-Robust Neural Aggregation Network for Video Face Identification. In 2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings (pp. 1675-1679). Article 8803028 (Proceedings - International Conference on Image Processing, ICIP; Vol. 2019-September). IEEE Computer Society. https://doi.org/10.1109/ICIP.2019.8803028

@inproceedings{e362477fceb7493c8646c73aae517467,

title = "Outlier-Robust Neural Aggregation Network for Video Face Identification",

abstract = "Current approaches for video face recognition rely on image sets containing faces of exclusively one identity. However, as image sets are created by unsupervised methods, it is necessary to consider outlier-afflicted sets for real-life applications. In this paper, we propose an Outlier-Robust Neural Aggregation Network (ORNAN). First, we embed each image into a feature space using a Convolutional Neural Network (CNN). With the help of two cascaded attention blocks, we predict outliers within the image set. By integrating this knowledge into our aggregation network, we adaptively aggregate all feature vectors to form a single feature, mitigating the influence of outliers and noisy features. We show that our network is robust against outliers using outlier-afflicted IJB-B and IJB-C benchmarks while maintaining similar performance without outliers.",

keywords = "biometrics, face identification, feature aggregation, video face recognition",

author = "Stefan Hormann and Martin Knoche and Maryam Babaee and Okan Kopuklu and Gerhard Rigoll",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 26th IEEE International Conference on Image Processing, ICIP 2019 ; Conference date: 22-09-2019 Through 25-09-2019",

year = "2019",

month = sep,

doi = "10.1109/ICIP.2019.8803028",

language = "English",

series = "Proceedings - International Conference on Image Processing, ICIP",

publisher = "IEEE Computer Society",

pages = "1675--1679",

booktitle = "2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings",

}

Hormann, S, Knoche, M, Babaee, M, Kopuklu, O & Rigoll, G 2019, Outlier-Robust Neural Aggregation Network for Video Face Identification. in 2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings., 8803028, Proceedings - International Conference on Image Processing, ICIP, vol. 2019-September, IEEE Computer Society, pp. 1675-1679, 26th IEEE International Conference on Image Processing, ICIP 2019, Taipei, Taiwan, Province of China, 22/09/19. https://doi.org/10.1109/ICIP.2019.8803028

Outlier-Robust Neural Aggregation Network for Video Face Identification. / Hormann, Stefan; Knoche, Martin; Babaee, Maryam et al.
2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings. IEEE Computer Society, 2019. p. 1675-1679 8803028 (Proceedings - International Conference on Image Processing, ICIP; Vol. 2019-September).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Outlier-Robust Neural Aggregation Network for Video Face Identification

AU - Hormann, Stefan

AU - Knoche, Martin

AU - Babaee, Maryam

AU - Kopuklu, Okan

AU - Rigoll, Gerhard

PY - 2019/9

Y1 - 2019/9

N2 - Current approaches for video face recognition rely on image sets containing faces of exclusively one identity. However, as image sets are created by unsupervised methods, it is necessary to consider outlier-afflicted sets for real-life applications. In this paper, we propose an Outlier-Robust Neural Aggregation Network (ORNAN). First, we embed each image into a feature space using a Convolutional Neural Network (CNN). With the help of two cascaded attention blocks, we predict outliers within the image set. By integrating this knowledge into our aggregation network, we adaptively aggregate all feature vectors to form a single feature, mitigating the influence of outliers and noisy features. We show that our network is robust against outliers using outlier-afflicted IJB-B and IJB-C benchmarks while maintaining similar performance without outliers.

AB - Current approaches for video face recognition rely on image sets containing faces of exclusively one identity. However, as image sets are created by unsupervised methods, it is necessary to consider outlier-afflicted sets for real-life applications. In this paper, we propose an Outlier-Robust Neural Aggregation Network (ORNAN). First, we embed each image into a feature space using a Convolutional Neural Network (CNN). With the help of two cascaded attention blocks, we predict outliers within the image set. By integrating this knowledge into our aggregation network, we adaptively aggregate all feature vectors to form a single feature, mitigating the influence of outliers and noisy features. We show that our network is robust against outliers using outlier-afflicted IJB-B and IJB-C benchmarks while maintaining similar performance without outliers.

KW - biometrics

KW - face identification

KW - feature aggregation

KW - video face recognition

UR - http://www.scopus.com/inward/record.url?scp=85076806570&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2019.8803028

DO - 10.1109/ICIP.2019.8803028

M3 - Conference contribution

AN - SCOPUS:85076806570

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 1675

EP - 1679

BT - 2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings

PB - IEEE Computer Society

T2 - 26th IEEE International Conference on Image Processing, ICIP 2019

Y2 - 22 September 2019 through 25 September 2019

ER -

Outlier-Robust Neural Aggregation Network for Video Face Identification

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this