Large Scale Hierarchical Anomaly Detection and Temporal Localization

Soumil Kanwal; Vineet Mehta; Abhinav Dhall

doi:10.1145/3394171.3416302

Large Scale Hierarchical Anomaly Detection and Temporal Localization

Soumil Kanwal, Vineet Mehta, Abhinav Dhall

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

2 Citations (Scopus)

Abstract

Abnormal event detection is a non-trivial task in machine learning. The primary reason behind this is that the abnormal class occurs sparsely, and its temporal location may not be available. In this paper, we propose a multiple feature-based approach for CitySCENE challenge-based anomaly detection. For motion and context information, Res3D and Res101 architectures are used. Object-level information is extracted by object detection feature-based pooling. Fusion of three channels above gives relatively high performance on the challenge Test set for the general anomaly task. We also show how our method can be used for temporal localisation of the abnormal activity event in a video.

Original language	English
Title of host publication	MM '20 - Proceedings of the 28th ACM International Conference on Multimedia
Place of Publication	New York, NY
Publisher	Association for Computing Machinery, Inc
Pages	4674-4678
Number of pages	5
ISBN (Electronic)	9781450379885
DOIs	https://doi.org/10.1145/3394171.3416302
Publication status	Published - 12 Oct 2020
Externally published	Yes
Event	28th ACM International Conference on Multimedia, MM 2020 - Virtual, Online, United States Duration: 12 Oct 2020 → 16 Oct 2020

Publication series

Name	Proceedings of the ACM International Conference on Multimedia
Volume	2020

Conference

Conference	28th ACM International Conference on Multimedia, MM 2020
Country/Territory	United States
City	Virtual, Online
Period	12/10/20 → 16/10/20

Bibliographical note

Publisher Copyright:
© 2020 ACM.

Keywords

anomaly detection
CitySCENE
convolutional neural networks

Access to Document

10.1145/3394171.3416302Licence: In Copyright

Cite this

@inproceedings{f56567355d014e87a9c1e56829205b3c,

title = "Large Scale Hierarchical Anomaly Detection and Temporal Localization",

abstract = "Abnormal event detection is a non-trivial task in machine learning. The primary reason behind this is that the abnormal class occurs sparsely, and its temporal location may not be available. In this paper, we propose a multiple feature-based approach for CitySCENE challenge-based anomaly detection. For motion and context information, Res3D and Res101 architectures are used. Object-level information is extracted by object detection feature-based pooling. Fusion of three channels above gives relatively high performance on the challenge Test set for the general anomaly task. We also show how our method can be used for temporal localisation of the abnormal activity event in a video.",

keywords = "anomaly detection, CitySCENE, convolutional neural networks",

author = "Soumil Kanwal and Vineet Mehta and Abhinav Dhall",

note = "Publisher Copyright: {\textcopyright} 2020 ACM.; 28th ACM International Conference on Multimedia, MM 2020 ; Conference date: 12-10-2020 Through 16-10-2020",

year = "2020",

month = oct,

day = "12",

doi = "10.1145/3394171.3416302",

language = "English",

series = "Proceedings of the ACM International Conference on Multimedia",

publisher = "Association for Computing Machinery, Inc",

pages = "4674--4678",

booktitle = "MM '20 - Proceedings of the 28th ACM International Conference on Multimedia",

}

Kanwal, S, Mehta, V & Dhall, A 2020, Large Scale Hierarchical Anomaly Detection and Temporal Localization. in MM '20 - Proceedings of the 28th ACM International Conference on Multimedia. Proceedings of the ACM International Conference on Multimedia, vol. 2020, Association for Computing Machinery, Inc, New York, NY, pp. 4674-4678, 28th ACM International Conference on Multimedia, MM 2020, Virtual, Online, United States, 12/10/20. https://doi.org/10.1145/3394171.3416302

Large Scale Hierarchical Anomaly Detection and Temporal Localization. / Kanwal, Soumil; Mehta, Vineet; Dhall, Abhinav.
MM '20 - Proceedings of the 28th ACM International Conference on Multimedia. New York, NY: Association for Computing Machinery, Inc, 2020. p. 4674-4678 (Proceedings of the ACM International Conference on Multimedia; Vol. 2020).

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

TY - GEN

T1 - Large Scale Hierarchical Anomaly Detection and Temporal Localization

AU - Kanwal, Soumil

AU - Mehta, Vineet

AU - Dhall, Abhinav

PY - 2020/10/12

Y1 - 2020/10/12

N2 - Abnormal event detection is a non-trivial task in machine learning. The primary reason behind this is that the abnormal class occurs sparsely, and its temporal location may not be available. In this paper, we propose a multiple feature-based approach for CitySCENE challenge-based anomaly detection. For motion and context information, Res3D and Res101 architectures are used. Object-level information is extracted by object detection feature-based pooling. Fusion of three channels above gives relatively high performance on the challenge Test set for the general anomaly task. We also show how our method can be used for temporal localisation of the abnormal activity event in a video.

AB - Abnormal event detection is a non-trivial task in machine learning. The primary reason behind this is that the abnormal class occurs sparsely, and its temporal location may not be available. In this paper, we propose a multiple feature-based approach for CitySCENE challenge-based anomaly detection. For motion and context information, Res3D and Res101 architectures are used. Object-level information is extracted by object detection feature-based pooling. Fusion of three channels above gives relatively high performance on the challenge Test set for the general anomaly task. We also show how our method can be used for temporal localisation of the abnormal activity event in a video.

KW - anomaly detection

KW - CitySCENE

KW - convolutional neural networks

UR - http://www.scopus.com/inward/record.url?scp=85106959363&partnerID=8YFLogxK

U2 - 10.1145/3394171.3416302

DO - 10.1145/3394171.3416302

M3 - Conference contribution

AN - SCOPUS:85106959363

T3 - Proceedings of the ACM International Conference on Multimedia

SP - 4674

EP - 4678

BT - MM '20 - Proceedings of the 28th ACM International Conference on Multimedia

PB - Association for Computing Machinery, Inc

CY - New York, NY

T2 - 28th ACM International Conference on Multimedia, MM 2020

Y2 - 12 October 2020 through 16 October 2020

ER -

Large Scale Hierarchical Anomaly Detection and Temporal Localization

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this