Puzzle-AE: Novelty Detection in Images through Solving Puzzles

Salehi, Mohammadreza; Eftekhar, Ainaz; Sadjadi, Niousha; Rohban, Mohammad Hossein; Rabiee, Hamid R.

doi:10.48550/arxiv.2008.12959

Cited by 11 publications

(14 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As in OOD detection, OSR, AD and ND, being robust against adversarial attacks is crucial. Recent works in OSR [133], [134], ND [18], [135], and OOD detection [136], [137] have investigated the effects of adversarial attacks on models. However more is needed.…”

Section: Adversarial Robustnessmentioning

confidence: 99%

“…From a completely different point of view, as it is mentioned in [140], adversarial robust training can be employed to boost learned feature space in a semantic way. This path has been followed in ARAE [18] and Puzzle-AE [135] to improve the performance of AEs in detecting unseen test time samples. Similar intention is followed in the one-class learning method [141] that shows robustness is beneficial for detecting novel samples.…”

Section: Adversarial Robustnessmentioning

confidence: 99%

See 1 more Smart Citation

A Unified Survey on Anomaly, Novelty, Open-Set, and Out-of-Distribution Detection: Solutions and Future Challenges

Salehi¹,

Mirzaei²,

Hendrycks³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Machine learning models often encounter samples that are diverged from the training distribution. Failure to recognize an out-of-distribution (OOD) sample, and consequently assign that sample to an in-class label significantly compromises the reliability of a model. The problem has gained significant attention due to its importance for safety deploying models in open-world settings. Detecting OOD samples is challenging due to the intractability of modeling all possible unknown distributions. To date, several research domains tackle the problem of detecting unfamiliar samples, including anomaly detection, novelty detection, one-class learning, open set recognition, and out-of-distribution detection. Despite having similar and shared concepts, out-of-distribution, open-set, and anomaly detection have been investigated independently. Accordingly, these research avenues have not cross-pollinated, creating research barriers. While some surveys intend to provide an overview of these approaches, they seem to only focus on a specific domain without examining the relationship between different domains. This survey aims to provide a cross-domain and comprehensive review of numerous eminent works in respective areas while identifying their commonalities. Researchers can benefit from the overview of research advances in different fields and develop future methodology synergistically. Furthermore, to the best of our knowledge, while there are surveys in anomaly detection or one-class learning, there is no comprehensive or up-to-date survey on out-of-distribution detection, which our survey covers extensively. Finally, having a unified cross-domain perspective, we discuss and shed light on future lines of research, intending to bring these fields closer together.

show abstract

Section: Adversarial Robustnessmentioning

confidence: 99%

Section: Adversarial Robustnessmentioning

confidence: 99%

A Unified Survey on Anomaly, Novelty, Open-Set, and Out-of-Distribution Detection: Solutions and Future Challenges

Salehi¹,

Mirzaei²,

Hendrycks³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Typically, some transformations for the input image are firstly performed, such as color removal or geometric transformations. Then the autoencoder is trained to reconstruct the original input image with the incomplete input image on which the transformation is made [31], [32]. This method can effectively increase the reconstruction difficulties for abnormal images, so their reconstruction errors are usually large.…”

Section: Image Reconstructionmentioning

confidence: 99%

Visual Anomaly Detection for Images: A Survey

Yang,

Xu,

et al. 2021

Preprint

View full text Add to dashboard Cite

Visual anomaly detection is an important and challenging problem in the field of machine learning and computer vision. This problem has attracted a considerable amount of attention in relevant research communities. Especially in recent years, the development of deep learning has sparked an increasing interest in the visual anomaly detection problem and brought a great variety of novel methods. In this paper, we provide a comprehensive survey of the classical and deep learning-based approaches for visual anomaly detection in the literature. We group the relevant approaches in view of their underlying principles and discuss their assumptions, advantages, and disadvantages carefully. We aim to help the researchers to understand the common principles of visual anomaly detection approaches and identify promising research directions in this field.

show abstract

“…Hopefully, the model learns meaningful features that reflect the normal nature of the data. The construction of an auxiliary task that motivates the model to learn these relevant features is not trivial, and several suggestions have been made such as geometric transformations classification [16], rotation classification [22], puzzlesolving [34] and CutPaste [23].…”

Section: Introductionmentioning

confidence: 99%

“…A number of papers have proposed applying transformations to the normal data and predicting which transformations have been applied. Predicting predefined geometric transformations [16], rotation [22], puzzle-solving [34] and CutPaste [23] are a few of the auxiliary tasks that have been suggested. In addition, (random) general transformations can be applied not only to images but also to tabular data, enabling anomalies to be detected in this domain as well [6].…”

Section: Introductionmentioning

confidence: 99%

Transformaly -- Two (Feature Spaces) Are Better Than One

Cohen¹,

Avidan²

2021

Preprint

View full text Add to dashboard Cite

Anomaly detection is a well-established research area that seeks to identify samples outside of a predetermined distribution. An anomaly detection pipeline is comprised of two main stages: (1) feature extraction and (2) normality score assignment. Recent papers used pre-trained networks for feature extraction achieving state-of-the-art results. However, the use of pre-trained networks does not fully-utilize the normal samples that are available at train time. This paper suggests taking advantage of this information by using teacher-student training. In our setting, a pretrained teacher network is used to train a student network on the normal training samples. Since the student network is trained only on normal samples, it is expected to deviate from the teacher network in abnormal cases. This difference can serve as a complementary representation to the pre-trained feature vector. Our method -T ransf ormaly -exploits a pre-trained Vision Transformer (ViT) to extract both feature vectors: the pre-trained (agnostic) features and the teacher-student (fine-tuned) features. We report stateof-the-art AUROC results in both the common unimodal setting, where one class is considered normal and the rest are considered abnormal, and the multimodal setting, where all classes but one are considered normal, and just one class is considered abnormal 1 .

show abstract

Puzzle-AE: Novelty Detection in Images through Solving Puzzles

Cited by 11 publications

References 0 publications

A Unified Survey on Anomaly, Novelty, Open-Set, and Out-of-Distribution Detection: Solutions and Future Challenges

A Unified Survey on Anomaly, Novelty, Open-Set, and Out-of-Distribution Detection: Solutions and Future Challenges

Visual Anomaly Detection for Images: A Survey

Transformaly -- Two (Feature Spaces) Are Better Than One

Contact Info

Product

Resources

About