Greedy vs. L1 Convex Optimization in Sparse Coding: Comparative Study in Abnormal Event Detection

Huamin Ren, Hong Pan, Søren Ingvor Olsen, Thomas B. Moeslund

Research output: Contribution to conference without publisher/journalPaper without publisher/journalResearchpeer-review

473 Downloads (Pure)

Abstract

Sparse representation has been applied successfully in many image analysis applications, including abnormal event detection, in which a baseline is to learn a dictionary from the training data and detect anomalies from its sparse codes. During this procedure, sparse codes which can be achieved through finding the L0-norm solution of the problem: min ||Y - D \alpha||_0, is crucial. Note that D refers to the dictionary and refers to the sparse codes. This L0-norm solution, however, is known as a NP-hard problem. Despite of the research achievements in some classification fields, such as face and action recognition, a comparative study of codes in abnormal event detection is less studied and hence no conclusion is gained on the effect of codes in detecting abnormalities. We constrict our comparison in two types of the above L0-norm solutions: greedy algorithms and convex L1-norm solutions. Considering the property of abnormal event detection, i.e., only normal videos are used as training data due to practical reasons, effective codes in classification application may not perform well in abnormality detection. Therefore, we compare the sparse codes and comprehensively evaluate their performance from various aspects to better understand their applicability, including computation time, reconstruction error, sparsity, detection accuracy on the UCSD Anomaly Dataset. Experiments show that greedy algorithms, especially MP and StOMP algorithm could achieve better abnormality detection with relatively less computations.

Original languageEnglish
Publication dateJul 2015
Number of pages6
Publication statusPublished - Jul 2015
EventICML '15 Workshop: FEAST 2015: ICML Workshop on Features and Structures - Lille Grand Palais, Lille, France
Duration: 6 Jul 201511 Jul 2015

Conference

ConferenceICML '15 Workshop
LocationLille Grand Palais
CountryFrance
CityLille
Period06/07/201511/07/2015

Fingerprint

Convex optimization
Glossaries
Error detection
Image analysis
Computational complexity
Experiments

Cite this

Ren, H., Pan, H., Olsen, S. I., & Moeslund, T. B. (2015). Greedy vs. L1 Convex Optimization in Sparse Coding: Comparative Study in Abnormal Event Detection. Paper presented at ICML '15 Workshop, Lille, France.
Ren, Huamin ; Pan, Hong ; Olsen, Søren Ingvor ; Moeslund, Thomas B. / Greedy vs. L1 Convex Optimization in Sparse Coding : Comparative Study in Abnormal Event Detection. Paper presented at ICML '15 Workshop, Lille, France.6 p.
@conference{0deaab927b254946a133a01f52650d8f,
title = "Greedy vs. L1 Convex Optimization in Sparse Coding: Comparative Study in Abnormal Event Detection",
abstract = "Sparse representation has been applied successfully in many image analysis applications, including abnormal event detection, in which a baseline is to learn a dictionary from the training data and detect anomalies from its sparse codes. During this procedure, sparse codes which can be achieved through finding the L0-norm solution of the problem: min ||Y - D \alpha||_0, is crucial. Note that D refers to the dictionary and refers to the sparse codes. This L0-norm solution, however, is known as a NP-hard problem. Despite of the research achievements in some classification fields, such as face and action recognition, a comparative study of codes in abnormal event detection is less studied and hence no conclusion is gained on the effect of codes in detecting abnormalities. We constrict our comparison in two types of the above L0-norm solutions: greedy algorithms and convex L1-norm solutions. Considering the property of abnormal event detection, i.e., only normal videos are used as training data due to practical reasons, effective codes in classification application may not perform well in abnormality detection. Therefore, we compare the sparse codes and comprehensively evaluate their performance from various aspects to better understand their applicability, including computation time, reconstruction error, sparsity, detection accuracy on the UCSD Anomaly Dataset. Experiments show that greedy algorithms, especially MP and StOMP algorithm could achieve better abnormality detection with relatively less computations.",
author = "Huamin Ren and Hong Pan and Olsen, {S{\o}ren Ingvor} and Moeslund, {Thomas B.}",
year = "2015",
month = "7",
language = "English",
note = "ICML '15 Workshop : FEAST 2015: ICML Workshop on Features and Structures ; Conference date: 06-07-2015 Through 11-07-2015",

}

Ren, H, Pan, H, Olsen, SI & Moeslund, TB 2015, 'Greedy vs. L1 Convex Optimization in Sparse Coding: Comparative Study in Abnormal Event Detection' Paper presented at ICML '15 Workshop, Lille, France, 06/07/2015 - 11/07/2015, .

Greedy vs. L1 Convex Optimization in Sparse Coding : Comparative Study in Abnormal Event Detection. / Ren, Huamin; Pan, Hong; Olsen, Søren Ingvor; Moeslund, Thomas B.

2015. Paper presented at ICML '15 Workshop, Lille, France.

Research output: Contribution to conference without publisher/journalPaper without publisher/journalResearchpeer-review

TY - CONF

T1 - Greedy vs. L1 Convex Optimization in Sparse Coding

T2 - Comparative Study in Abnormal Event Detection

AU - Ren, Huamin

AU - Pan, Hong

AU - Olsen, Søren Ingvor

AU - Moeslund, Thomas B.

PY - 2015/7

Y1 - 2015/7

N2 - Sparse representation has been applied successfully in many image analysis applications, including abnormal event detection, in which a baseline is to learn a dictionary from the training data and detect anomalies from its sparse codes. During this procedure, sparse codes which can be achieved through finding the L0-norm solution of the problem: min ||Y - D \alpha||_0, is crucial. Note that D refers to the dictionary and refers to the sparse codes. This L0-norm solution, however, is known as a NP-hard problem. Despite of the research achievements in some classification fields, such as face and action recognition, a comparative study of codes in abnormal event detection is less studied and hence no conclusion is gained on the effect of codes in detecting abnormalities. We constrict our comparison in two types of the above L0-norm solutions: greedy algorithms and convex L1-norm solutions. Considering the property of abnormal event detection, i.e., only normal videos are used as training data due to practical reasons, effective codes in classification application may not perform well in abnormality detection. Therefore, we compare the sparse codes and comprehensively evaluate their performance from various aspects to better understand their applicability, including computation time, reconstruction error, sparsity, detection accuracy on the UCSD Anomaly Dataset. Experiments show that greedy algorithms, especially MP and StOMP algorithm could achieve better abnormality detection with relatively less computations.

AB - Sparse representation has been applied successfully in many image analysis applications, including abnormal event detection, in which a baseline is to learn a dictionary from the training data and detect anomalies from its sparse codes. During this procedure, sparse codes which can be achieved through finding the L0-norm solution of the problem: min ||Y - D \alpha||_0, is crucial. Note that D refers to the dictionary and refers to the sparse codes. This L0-norm solution, however, is known as a NP-hard problem. Despite of the research achievements in some classification fields, such as face and action recognition, a comparative study of codes in abnormal event detection is less studied and hence no conclusion is gained on the effect of codes in detecting abnormalities. We constrict our comparison in two types of the above L0-norm solutions: greedy algorithms and convex L1-norm solutions. Considering the property of abnormal event detection, i.e., only normal videos are used as training data due to practical reasons, effective codes in classification application may not perform well in abnormality detection. Therefore, we compare the sparse codes and comprehensively evaluate their performance from various aspects to better understand their applicability, including computation time, reconstruction error, sparsity, detection accuracy on the UCSD Anomaly Dataset. Experiments show that greedy algorithms, especially MP and StOMP algorithm could achieve better abnormality detection with relatively less computations.

M3 - Paper without publisher/journal

ER -

Ren H, Pan H, Olsen SI, Moeslund TB. Greedy vs. L1 Convex Optimization in Sparse Coding: Comparative Study in Abnormal Event Detection. 2015. Paper presented at ICML '15 Workshop, Lille, France.