Multimodal subspace support vector data description

Sohrab, Fahad; Raitoharju, Jenni; Iosifidis, Alexandros; Gabbouj, Moncef

doi:10.1016/j.patcog.2020.107648

Cited by 27 publications

(9 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These two modules can be improved furthermore. For example, it is interesting to leverage effective one-class classification methods such as SVDD [14], subspace SVDD [37] and multimodal subspace SVDD [38]. This is a new topic under our current investigation.…”

Section: Discussionmentioning

confidence: 99%

AnomalyHop: An SSL-based Image Anomaly Localization Method

Zhang

Wang²,

Wang³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

An image anomaly localization method based on the successive subspace learning (SSL) framework, called Anomaly-Hop, is proposed in this work. AnomalyHop consists of three modules: 1) feature extraction via successive subspace learning (SSL), 2) normality feature distributions modeling via Gaussian models, and 3) anomaly map generation and fusion. Comparing with state-of-the-art image anomaly localization methods based on deep neural networks (DNNs), AnomalyHop is mathematically transparent, easy to train, and fast in its inference speed. Besides, its area under the ROC curve (ROC-AUC) performance on the MVTec AD dataset is 95.9%, which is among the best of several benchmarking methods. 1

show abstract

Section: Discussionmentioning

confidence: 99%

AnomalyHop: An SSL-based Image Anomaly Localization Method

Zhang

Wang²,

Wang³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…For the RBF kernel in (6), the corresponding probability distribution p κ (•) is N (0, σ 2 I) [34] and, thus, random Fourier features can be generated by sampling vectors η 1 , . .…”

Section: B Kernel-based Methodsmentioning

confidence: 99%

“…Traditional well-known one-class classification techniques include One-class Support Vector Machine (OC-SVM) [2] and Support Vector Data Description (SVDD) [3], which are commonly applied as non-linear models exploiting the kernel trick. Various extensions of both methods have been proposed (e.g., [4]- [6]), and recently also deep neural network-based variants have been proposed [7]- [9]. Some recent works have shown that also the traditional methods used on top of deep features may be useful.…”

Section: Introductionmentioning

confidence: 99%

Generalized Reference Kernel for One-class Classification

Raitoharju¹,

Iosifidis²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

In this paper, we formulate a new generalized reference kernel hoping to improve the original base kernel using a set of reference vectors. Depending on the selected reference vectors, our formulation shows similarities to approximate kernels, random mappings, and Non-linear Projection Trick. Focusing on small-scale one-class classification, our analysis and experimental results show that the new formulation provides approaches to regularize, adjust the rank, and incorporate additional information into the kernel itself, leading to improved one-class classification accuracy.

show abstract

“…Contrary to class-specific algorithms, the OCC models do not require information from negative samples during training. For the OCC model, we propose using the Multi-modal Subspace Support Vector Data Description (MS-SVDD) [21] due to its feasibility with multi-view echocardiographic data. MS-SVDD maps the multi-view feature vectors to a lower-dimensional optimized feature space shared by features from different views of echocardiography as illustrated in Figure 1.…”

Section: Methodsmentioning

confidence: 99%

Early Myocardial Infarction Detection with One-Class Classification over Multi-view Echocardiography

Degerli¹,

Sohrab²,

Kıranyaz³

et al. 2022

Computing in Cardiology Conference (CinC)

View full text Add to dashboard Cite

Myocardial infarction (MI) is the leading cause of mortality and morbidity in the world. Early therapeutics of MI can ensure the prevention of further myocardial necrosis. Echocardiography is the fundamental imaging technique that can reveal the earliest sign of MI. However, the scarcity of echocardiographic datasets for the MI detection is the major issue for training data-driven classification algorithms. In this study, we propose a framework for early detection of MI over multi-view echocardiography that leverages one-class classification (OCC) techniques. The OCC techniques are used to train a model for detecting a specific target class using instances from that particular category only. We investigated the usage of uni-modal and multi-modal one-class classification techniques in the proposed framework using the HMC-QU dataset that includes apical 4-chamber (A4C) and apical 2-chamber (A2C) views in a total of 260 echocardiography recordings. Experimental results show that the multimodal approach achieves a sensitivity level of 85.23% and F1-Score of 80.21%.

show abstract

Multimodal subspace support vector data description

Cited by 27 publications

References 25 publications

AnomalyHop: An SSL-based Image Anomaly Localization Method

AnomalyHop: An SSL-based Image Anomaly Localization Method

Generalized Reference Kernel for One-class Classification

Early Myocardial Infarction Detection with One-Class Classification over Multi-view Echocardiography

Contact Info

Product

Resources

About