Anomaly Detection Using GANs for Visual Inspection in Noisy Training Data

Kimura, Masanari; Yanagihara, Takashi

doi:10.48550/arxiv.1807.01136

Cited by 5 publications

(11 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, the growth of outliers has a negligible effect on the performance of our proposed method and thus it does not experience a noticeable drop in F 1 -score measure. Results on Caltech-256: We compare our work with other six methods therein designed specifically for outlier detection, including R-graph [42], REAPER [14], OutlierPursuit [40], LRR [17], SSGAN [12] and ALOCC [33]. The performance metrics of this experiment are F 1 -score and the Table 2: Results on Caltech-256 dataset.…”

Section: Image Outlier Detectionmentioning

confidence: 99%

“…Two first rows: Inliers are from one category of images, with 50% portion of outliers; Two second rows: Inliers are from three categories of images, with 50% portion of outliers; Two last rows: Inliers come from five categories of images, while outliers form 50% of the samples. R-graph [42] REAPER [14] OutlierPursuit [40] LRR [17] SSGAN [12] ALOCC [33] Ours…”

Section: Image Outlier Detectionmentioning

confidence: 99%

See 1 more Smart Citation

G2D: Generate to Detect Anomaly

Pourreza¹,

Mohammadi²,

Khaki³

et al. 2020

Preprint

View full text Add to dashboard Cite

In this paper, we propose a novel method for irregularity detection. Previous researches solve this problem as a One-Class Classification (OCC) task where they train a reference model on all of the available samples. Then, they consider a test sample as an anomaly if it has a diversion from the reference model. Generative Adversarial Networks (GANs) have achieved the most promising results for OCC while implementing and training such networks, especially for the OCC task, is a cumbersome and computationally expensive procedure. To cope with the mentioned challenges, we present a simple but effective method to solve the irregularity detection as a binary classification task in order to make the implementation easier along with improving the detection performance. We learn two deep neural networks (generator and discriminator) in a GAN-style setting on merely the normal samples. During training, the generator gradually becomes an expert to generate samples which are similar to the normal ones. In the training phase, when the generator fails to produce normal data (in the early stages of learning and also prior to the complete convergence), it can be considered as an irregularity generator. In this way, we simultaneously generate the irregular samples. Afterward, we train a binary classifier on the generated anomalous samples along with the normal instances in order to be capable of detecting irregularities. The proposed framework applies to different related applications of outlier and anomaly detection in images and videos, respectively. The results confirm that our proposed method is superior to the baseline and state-of-the-art solutions.

show abstract

Section: Image Outlier Detectionmentioning

confidence: 99%

Section: Image Outlier Detectionmentioning

confidence: 99%

G2D: Generate to Detect Anomaly

Pourreza¹,

Mohammadi²,

Khaki³

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…The detection of anomaly areas in images is widely studied [1,5,2,3,7]. The VAEs-based method finds anomaly areas by taking the difference between the input image and the reconstructed image.…”

Section: Related Workmentioning

confidence: 99%

“…With the trained generator G, if a given new query image x is from the normal data distribution, noise ẑ must exist in the latent space where G(ẑ) becomes identical to x. Furthermore, a method that improves AnoGAN has also been proposed [3,8].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

PNUNet: Anomaly Detection using Positive-and-Negative Noise based on Self-Training Procedure

Kimura¹

2019

Preprint

Self Cite

View full text Add to dashboard Cite

We propose the novel framework for anomaly detection in images. Our new framework, PNUNet, is based on many normal data and few anomalous data. We assume that some noises are added to the input images and learn to remove the noise. In addition, the proposed method achieves significant performance improvement by updating the noise assumed in the inputs using a self-training framework. The experimental results for the benchmark datasets show the usefulness of our new anomaly detection framework.

show abstract

Anomaly Detection Based on Zero-Shot Outlier Synthesis and Hierarchical Feature Distillation

Rivera

Khan

Bekkouch

et al. 2022

IEEE Trans. Neural Netw. Learning Syst.

View full text Add to dashboard Cite

Anomaly detection suffers from unbalanced data since anomalies are quite rare. Synthetically generated anomalies are a solution to such ill or not fully defined data. However, synthesis requires an expressive representation to guarantee the quality of the generated data. In this paper, we propose a two-level hierarchical latent space representation that distills inliers' feature-descriptors (through autoencoders) into more robust representations based on a variational family of distributions (through a variational autoencoder) for zero-shot anomaly generation. From the learned latent distributions, we select those that lie on the outskirts of the training data as synthetic-outlier generators. And, we synthesize from them, i.e., generate negative samples without seen them before, to train binary classifiers. We found that the use of the proposed hierarchical structure for feature distillation and fusion creates robust and general representations that allow us to synthesize pseudo outlier samples. And in turn, train robust binary classifiers for true outlier detection (without the need for actual outliers during training). We demonstrate the performance of our proposal on several benchmarks for anomaly detection.

show abstract

Anomaly Detection Using GANs for Visual Inspection in Noisy Training Data

Cited by 5 publications

References 0 publications

G2D: Generate to Detect Anomaly

G2D: Generate to Detect Anomaly

PNUNet: Anomaly Detection using Positive-and-Negative Noise based on Self-Training Procedure

Anomaly Detection Based on Zero-Shot Outlier Synthesis and Hierarchical Feature Distillation

Contact Info

Product

Resources

About