Automated Testing of Graphics Units by Deep-Learning Detection of Visual Anomalies

Lev Faivishevsky; Adi Szeskin; Ashwin K. Muppalla; Ravid Shwartz-Ziv; Itamar Ben Ari; Ronen Laperdon; Benjamin Melloul; Tahi Hollander; Tom Hope; Amitai Armon

Automated Testing of Graphics Units by Deep-Learning Detection of Visual Anomalies

Lev Faivishevsky, Adi Szeskin, Ashwin K. Muppalla, Ravid Shwartz-Ziv, Itamar Ben Ari, Ronen Laperdon, Benjamin Melloul, Tahi Hollander, Tom Hope, Amitai Armon

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

Abstract

We present a novel system for performing real-time detection of diverse visual corruptions in videos, for validating the quality of graphics units in our company. The system is used for several types of content, including movies and 3D graphics, with strict constraints on low false alert rates and real-time processing of millions of video frames per day. These constraints required novel solutions involving both hardware and software, including new supervised and weakly-supervised methods we developed. Our deployed system has enabled a 20X reduction of human effort and discovering new corruptions missed by humans and existing approaches.

Original language	English
Title of host publication	KDD 2021 - Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
Publisher	Association for Computing Machinery
Pages	2811-2821
Number of pages	11
ISBN (Electronic)	9781450383325
State	Published - 14 Aug 2021
Externally published	Yes
Event	27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD 2021 - Virtual, Online, Singapore Duration: 14 Aug 2021 → 18 Aug 2021

Publication series

Name	Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Conference

Conference	27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD 2021
Country/Territory	Singapore
City	Virtual, Online
Period	14/08/21 → 18/08/21

Bibliographical note

Publisher Copyright:
© 2021 ACM.

Keywords

anomaly detection
computer vision
deep learning
graphics processors validation
multiple instance learning

Cite this

Faivishevsky, L., Szeskin, A., Muppalla, A. K., Shwartz-Ziv, R., Ben Ari, I., Laperdon, R., Melloul, B., Hollander, T., Hope, T., & Armon, A. (2021). Automated Testing of Graphics Units by Deep-Learning Detection of Visual Anomalies. In KDD 2021 - Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (pp. 2811-2821). (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining). Association for Computing Machinery.

Faivishevsky, Lev ; Szeskin, Adi ; Muppalla, Ashwin K. et al. / Automated Testing of Graphics Units by Deep-Learning Detection of Visual Anomalies. KDD 2021 - Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2021. pp. 2811-2821 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).

@inproceedings{121de096afd14aa8b0e2410e35d379ac,

title = "Automated Testing of Graphics Units by Deep-Learning Detection of Visual Anomalies",

abstract = "We present a novel system for performing real-time detection of diverse visual corruptions in videos, for validating the quality of graphics units in our company. The system is used for several types of content, including movies and 3D graphics, with strict constraints on low false alert rates and real-time processing of millions of video frames per day. These constraints required novel solutions involving both hardware and software, including new supervised and weakly-supervised methods we developed. Our deployed system has enabled a 20X reduction of human effort and discovering new corruptions missed by humans and existing approaches.",

keywords = "anomaly detection, computer vision, deep learning, graphics processors validation, multiple instance learning",

author = "Lev Faivishevsky and Adi Szeskin and Muppalla, {Ashwin K.} and Ravid Shwartz-Ziv and {Ben Ari}, Itamar and Ronen Laperdon and Benjamin Melloul and Tahi Hollander and Tom Hope and Amitai Armon",

note = "Publisher Copyright: {\textcopyright} 2021 ACM.; 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD 2021 ; Conference date: 14-08-2021 Through 18-08-2021",

year = "2021",

month = aug,

day = "14",

language = "אנגלית",

series = "Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

publisher = "Association for Computing Machinery",

pages = "2811--2821",

booktitle = "KDD 2021 - Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining",

}

Faivishevsky, L, Szeskin, A, Muppalla, AK, Shwartz-Ziv, R, Ben Ari, I, Laperdon, R, Melloul, B, Hollander, T, Hope, T & Armon, A 2021, Automated Testing of Graphics Units by Deep-Learning Detection of Visual Anomalies. in KDD 2021 - Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Association for Computing Machinery, pp. 2811-2821, 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD 2021, Virtual, Online, Singapore, 14/08/21.

Automated Testing of Graphics Units by Deep-Learning Detection of Visual Anomalies. / Faivishevsky, Lev; Szeskin, Adi; Muppalla, Ashwin K. et al.
KDD 2021 - Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2021. p. 2811-2821 (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).

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

TY - GEN

T1 - Automated Testing of Graphics Units by Deep-Learning Detection of Visual Anomalies

AU - Faivishevsky, Lev

AU - Szeskin, Adi

AU - Muppalla, Ashwin K.

AU - Shwartz-Ziv, Ravid

AU - Ben Ari, Itamar

AU - Laperdon, Ronen

AU - Melloul, Benjamin

AU - Hollander, Tahi

AU - Hope, Tom

AU - Armon, Amitai

PY - 2021/8/14

Y1 - 2021/8/14

N2 - We present a novel system for performing real-time detection of diverse visual corruptions in videos, for validating the quality of graphics units in our company. The system is used for several types of content, including movies and 3D graphics, with strict constraints on low false alert rates and real-time processing of millions of video frames per day. These constraints required novel solutions involving both hardware and software, including new supervised and weakly-supervised methods we developed. Our deployed system has enabled a 20X reduction of human effort and discovering new corruptions missed by humans and existing approaches.

AB - We present a novel system for performing real-time detection of diverse visual corruptions in videos, for validating the quality of graphics units in our company. The system is used for several types of content, including movies and 3D graphics, with strict constraints on low false alert rates and real-time processing of millions of video frames per day. These constraints required novel solutions involving both hardware and software, including new supervised and weakly-supervised methods we developed. Our deployed system has enabled a 20X reduction of human effort and discovering new corruptions missed by humans and existing approaches.

KW - anomaly detection

KW - computer vision

KW - deep learning

KW - graphics processors validation

KW - multiple instance learning

UR - http://www.scopus.com/inward/record.url?scp=85114936371&partnerID=8YFLogxK

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85114936371

T3 - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

SP - 2811

EP - 2821

BT - KDD 2021 - Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

PB - Association for Computing Machinery

T2 - 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD 2021

Y2 - 14 August 2021 through 18 August 2021

ER -

Faivishevsky L, Szeskin A, Muppalla AK, Shwartz-Ziv R, Ben Ari I, Laperdon R et al. Automated Testing of Graphics Units by Deep-Learning Detection of Visual Anomalies. In KDD 2021 - Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery. 2021. p. 2811-2821. (Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining).

Automated Testing of Graphics Units by Deep-Learning Detection of Visual Anomalies

Abstract

Publication series

Conference

Bibliographical note

Keywords

Other files and links

Fingerprint

Cite this