Adversarial Discriminative Attention for Robust Anomaly Detection

Kimura, Daiki; Chaudhury, Subhajit; Narita, Minori; Munawar, Asim; Tachibana, Ryuki

doi:10.1109/wacv45572.2020.9093428

Cited by 33 publications

(17 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To validate the proposed method, we performed a categoryout experiment referring to image-based anomaly detection methods [12,13]. To verify the anomaly detection method for 3D point clouds, we conducted two different experiments:…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…We compared the proposed model with an anomaly detection result with FoldingNet and summarized the results in Table 1. Following the experimental setup in [12,13], we measured the average AUC by computing the area under the ROC with varying threshold values for the anomaly scores. We report the AUC performance of two of the models, with and without D KLrec in the loss.…”

Section: Quantitative Evaluationmentioning

confidence: 99%

“…In addition, we have conducted many experiments to evaluate the loss during training and the method for measuring the anomaly score A(x) of 3D point clouds, which is appropriate for solving the task of anomaly detection in 3D point clouds. To validate the proposed method, we performed a category-out experiment on the ShapeNet dataset [17], referring to the experiments of anomaly detection methods on images [12,13].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Toward Unsupervised 3d Point Cloud Anomaly Detection Using Variational Autoencoder

Masuda

Hachiuma

Fujii

et al. 2021

2021 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

In this paper, we present an end-to-end unsupervised anomaly detection framework for 3D point clouds. To the best of our knowledge, this is the first work to tackle the anomaly detection task on a general object represented by a 3D point cloud. We propose a deep variational autoencoder based unsupervised anomaly detection network adapted to the 3D point cloud and an anomaly score specifically for 3D point clouds.To verify the effectiveness of the model, we conducted extensive experiments on ShapeNet dataset. Through quantitative and qualitative evaluation, we demonstrate that the proposed method outperforms the baseline method.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Quantitative Evaluationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Toward Unsupervised 3d Point Cloud Anomaly Detection Using Variational Autoencoder

Masuda

Hachiuma

Fujii

et al. 2021

2021 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

show abstract

“…This is the first work that develops a methodology combining negative training using the learned distribution boundary with a GAN to robustly detect few-shots of OoC. The evaluation shows that anomaly detection with Outlier Exposure combined with few-shot OoD detection [33], [35], [36], within the REFGAN methodology, is beneficial. In the few-shot setting [25], [1], first moving away from and then detecting FSOoC is effective.…”

Section: Discussionmentioning

confidence: 99%

Few-Shot Adaptive Detection of Objects of Concern Using Generative Models with Negative Retraining

Dionelis

Vaighan

Tsaftaris

2021

2021 IEEE 33rd International Conference on Tools With Artificial Intelligence (ICTAI)

View full text Add to dashboard Cite

Detecting objects which we are interested in, Objects of Concern (OoC), is nowadays attracting attention. In aviation and transport, it is important to robustly detect OoC in security images. OoC are rare, differ from typical samples, and may be unknown during training. Most OoC detection methods need to be trained on large datasets to achieve good performance and have limited real-world generalization ability. A large variety of samples is needed, and it is expensive to collect and label large datasets due to the rarity of OoC. To address such limitations, we propose the negative REtraining with Few-shots Generative Adversarial Network (REFGAN) for detecting OoC. REFGAN aims at automatically identifying OoC by learning from Objects of No Concern (OoNC) and OoC. Our methodology comprises learning a prior using OoNC, and few-shot adaptation using the Few-Shot OoC (FSOoC). We propose a methodology to robustly perform few-shot adaptive detection of OoC using GANs and learned distribution boundaries. The evaluation of REFGAN on the Baggage SIXray dataset shows that when FSOoC are used, our model outperforms the prior improving OoC detection, and outperforms recent benchmarks by approximately 6.3% in mean values. REFGAN using few-shots of 80 samples shows a robust comparable performance compared to REFGAN using all the samples for retraining and model adaptation. REFGAN can detect unknown OoC and its evaluation on SIXray and CIFAR-10 shows robustness against the number of few-shot samples of OoC. REFGAN on CIFAR-10 outperforms benchmarks by approximately 16% using few-shots of 80 and of 10 samples.

show abstract

“…1(e) and (f), these methods approximately detect anomaly regions. In addition, in GradCAM-based methods, GradCAM (Selvaraju et al (2017)) is used to generate anomaly maps to detect regions that influence the decision of the trained model (Kimura et al (2020); Venkataramanan et al (2020)). CutPaste (Li et al (2021)) introduces a self-supervised framework using a simple effective augmentation that encourages the model to find local irregularities.…”

Section: Introductionmentioning

confidence: 99%

AnoSeg: Anomaly Segmentation Network Using Self-Supervised Learning

Song,

Kong,

Park

et al. 2021

Preprint

View full text Add to dashboard Cite

Anomaly segmentation, which localizes defective areas, is an important component in large-scale industrial manufacturing. However, most recent researches have focused on anomaly detection. This paper proposes a novel anomaly segmentation network (AnoSeg) that can directly generate an accurate anomaly map using self-supervised learning. For highly accurate anomaly segmentation, the proposed AnoSeg considers three novel techniques: Anomaly data generation based on hard augmentation, self-supervised learning with pixel-wise and adversarial losses, and coordinate channel concatenation. First, to generate synthetic anomaly images and reference masks for normal data, the proposed method uses hard augmentation to change the normal sample distribution. Then, the proposed AnoSeg is trained in a self-supervised learning manner from the synthetic anomaly data and normal data. Finally, the coordinate channel, which represents the pixel location information, is concatenated to an input of AnoSeg to consider the positional relationship of each pixel in the image. The estimated anomaly map can also be utilized to improve the performance of anomaly detection. Our experiments show that the proposed method outperforms the state-of-the-art anomaly detection and anomaly segmentation methods for the MVTec AD dataset. In addition, we compared the proposed method with the existing methods through the intersection over union (IoU) metric commonly used in segmentation tasks and demonstrated the superiority of our method for anomaly segmentation.

show abstract

Adversarial Discriminative Attention for Robust Anomaly Detection

Cited by 33 publications

References 27 publications

Toward Unsupervised 3d Point Cloud Anomaly Detection Using Variational Autoencoder

Toward Unsupervised 3d Point Cloud Anomaly Detection Using Variational Autoencoder

Few-Shot Adaptive Detection of Objects of Concern Using Generative Models with Negative Retraining

AnoSeg: Anomaly Segmentation Network Using Self-Supervised Learning

Contact Info

Product

Resources

About