Video anomaly detection and localization via multivariate gaussian fully convolution adversarial autoencoder

Li, Nanjun; Chang, Faliang

doi:10.1016/j.neucom.2019.08.044

Cited by 87 publications

(41 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, some of the reconstructionbased methods exploit the difference of latent representations between normal samples and abnormal samples to detect anomalies. Fan et al [7] and Li et al [8] used Variational Auto-Encoders (VAEs) to reconstruct input frames, and the distribution difference of latent representations was used to compute regularity scores.…”

Section: A Reconstruction-based Methodsmentioning

confidence: 99%

“…Thus, deep-learning-based semi-supervised anomaly detection methods have been proposed and achieved significant improvements. Generally, these methods can be devided into two categories: i) reconstruction-based methods [3]- [8]. Reconstruction-based methods believe that normal events can be reconstructed correctly by models trained with normality.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

TransAnomaly: Video Anomaly Detection Using Video Vision Transformer

Cai²,

Zhou

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Video anomaly detection is challenging because abnormal events are unbounded, rare, equivocal, irregular in real scenes. In recent years, transformers have demonstrated powerful modelling abilities for sequence data. Thus, we attempt to apply transformers to video anomaly detection. In this paper, we propose a prediction-based video anomaly detection approach named TransAnomaly. Our model combines the U-Net and the Video Vision Transformer (ViViT) to capture richer temporal information and more global contexts. To make full use of the ViViT for the prediction, we modified the ViViT to make it capable of video prediction. Experiments on benchmark datasets show that the addition of the transformer module improves the anomaly detection performance. In addition, we calculate regularity scores with sliding windows and evaluate the impact of different window sizes and strides. With proper settings, our model outperforms other state-of-the-art prediction-based video anomaly detection approaches. Furthermore, our model can perform anomaly localization by tracking the location of patches with lower regularity scores.

show abstract

Section: A Reconstruction-based Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

TransAnomaly: Video Anomaly Detection Using Video Vision Transformer

Cai²,

Zhou

et al. 2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The specialised networks are used in conjunction with autoencoders to infer the relationship between the neighbouring frames in a normal event as well as in an anomalous event. Li and Chang [26] implements a generative adversarial network for anomaly detection in surveillance videos. The idea is that the latent representation for the generator is based on the fact that they differ for normal and anomalous videos.…”

Section: Video Summarisation Based On Anomaly Detectionmentioning

confidence: 99%

An aggregated deep convolutional recurrent model for event based surveillance video summarisation: A supervised approach

Sreeja

Kovoor

2021

IET Computer Vision

View full text Add to dashboard Cite

Surveillance video summarisation is characterised by extracting video segments containing abnormal events from surveillance video footages. Accurate identification of abnormal events from surveillance footages is of paramount importance in surveillance video summarisation. Accordingly, the proposed framework builds an aggregated convolutional recurrent model that can precisely detect the suspicious events in a surveillance footage, by employing a supervised learning which is found to yield better results compared with unsupervised counterparts. The preliminary stage in the model is a multilayer Convolutional Neural Network for frame-level feature extraction followed by stacked bidirectional Gated Recurrent Unit for sequence-level feature extraction and classification. Since the video clips used for training are not implicit to surveillance, a block-based approach for testing on surveillance videos is proposed. The results evaluated on two custom datasets, Streets and Campus, prove that the proposed model produces remarkable results leveraging the properties of bidirectional GRU with supervised learning. Extensive experimental analysis on selection of optimum architecture is conducted which substantiates the significance of stacked bidirectional GRUs over unidirectional ones. Additionally, qualitative results ensure that summaries produced are concise, representative, complete, diverse and informative. Moreover, comparison of the performance of the proposed model with state of the art certainly proves the superiority of the proposed model.This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

show abstract

“…The proposed method consists of a feature extraction part and a detection part. The feature extraction is realized by Convolutional Adversarial Autoencoder (CAAE) [6,7] and Principal Component Analysis (PCA). CAAE is an extension of Adversarial Autoencoder (AAE) [8], where the layers of AAE are replaced with convolution and deconvolution layers.…”

Section: Introductionmentioning

confidence: 99%

Anomaly Detection Using Convolutional Adversarial Autoencoder and One-class SVM for Landslide Area Detection from Synthetic Aperture Radar Images

Mabu¹,

Hirata²,

Kuremoto³

2021

JRNAL

View full text Add to dashboard Cite

An anomaly detection model using deep learning for detecting disaster-stricken (landslide) areas in synthetic aperture radar images is proposed. Since it is difficult to obtain a large number of training images, especially disaster area images, with annotations, we design an anomaly detection model that only uses normal area images for the training, where the proposed model combines a convolutional adversarial autoencoder, principal component analysis, and one-class support vector machine.In the experiments, the ability in detecting normal and abnormal areas is evaluated.

show abstract

Video anomaly detection and localization via multivariate gaussian fully convolution adversarial autoencoder

Cited by 87 publications

References 18 publications

TransAnomaly: Video Anomaly Detection Using Video Vision Transformer

TransAnomaly: Video Anomaly Detection Using Video Vision Transformer

An aggregated deep convolutional recurrent model for event based surveillance video summarisation: A supervised approach

Anomaly Detection Using Convolutional Adversarial Autoencoder and One-class SVM for Landslide Area Detection from Synthetic Aperture Radar Images

Contact Info

Product

Resources

About