Detection of Video Anomalies Using Convolutional Autoencoders and One-Class Support Vector Machines

Gutoski, Matheus; Aquino, Nelson Marcelo Romero; Ribeiro, Matilde; Lazzaretti, André Engênio; Lopes, Heitor Silvério

doi:10.21528/cbic2017-49

Cited by 26 publications

(17 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They train the CAE on only normal frames and used a regularized reconstruction error as a score to identify normal and anomalous frames. They extend their work by using reconstruction error from the CAE trained on normal video frames as an input to a one-class SVM and showed similar results [10]. Tran and Hogg [28] use CAE and oneclass SVM for video anomaly detection.…”

Section: Related Workmentioning

confidence: 87%

DeepFall: Non-Invasive Fall Detection with Deep Spatio-Temporal Convolutional Autoencoders

Nogas

Khan

Mihailidis

2019

J Healthc Inform Res

View full text Add to dashboard Cite

Human falls rarely occur; however, detecting falls is very important from the health and safety perspective. Due to the rarity of falls, it is difficult to employ supervised classification techniques to detect them. Moreover, in these highly skewed situations it is also difficult to extract domain specific features to identify falls. In this paper, we present a novel framework, DeepFall, which formulates the fall detection problem as an anomaly detection problem. The DeepFall framework presents the novel use of deep spatio-temporal convolutional autoencoders to learn spatial and temporal features from normal activities using non-invasive sensing modalities. We also present a new anomaly scoring method that combines the reconstruction score of frames across a video sequences to detect unseen falls. We tested the DeepFall framework on three publicly available datasets collected through non-invasive sensing modalities, thermal camera and depth cameras and show superior results in comparison to traditional autoencoder and convolutional autoencoder methods to identify unseen falls.

show abstract

Section: Related Workmentioning

confidence: 87%

DeepFall: Non-Invasive Fall Detection with Deep Spatio-Temporal Convolutional Autoencoders

Nogas

Khan

Mihailidis

2019

J Healthc Inform Res

View full text Add to dashboard Cite

show abstract

“…[103] for anomaly detection for effective detection of anomalies. Many works have applied the deep learning models for video surveillance anomaly detection in [104]- [106].…”

Section: Application Domainsmentioning

confidence: 99%

Anomalous Example Detection in Deep Learning: A Survey

et al. 2020

View full text Add to dashboard Cite

Deep Learning (DL) is vulnerable to out-of-distribution and adversarial examples resulting in incorrect outputs. To make DL more robust, several posthoc (or runtime) anomaly detection techniques to detect (and discard) these anomalous samples have been proposed in the recent past. This survey tries to provide a structured and comprehensive overview of the research on anomaly detection for DL based applications. We provide a taxonomy for existing techniques based on their underlying assumptions and adopted approaches. We discuss various techniques in each of the categories and provide the relative strengths and weaknesses of the approaches. Our goal in this survey is to provide an easier yet better understanding of the techniques belonging to different categories in which research has been done on this topic. Finally, we highlight the unsolved research challenges while applying anomaly detection techniques in DL systems and present some high-impact future research directions.

show abstract

“…For example, autoencoders were used for target recognition using radar images in [47], and outlier removal in [48]. Autoencoders were also used to detect abnormalities in machines by detecting abnormal operation sounds [49], and to detect anomalies in video frames [50]. The idea is based on the premise that a trained autoencoder will output a low reconstruction error when the data it receives belongs to the same, or a similar distribution as the data used to train the model, but a high reconstruction error otherwise.…”

Section: Device Authentication With Dacmentioning

confidence: 99%

Device Authentication Codes based on RF Fingerprinting using Deep Learning

Bassey¹,

Li²,

Qian³

2021

ICST Transactions on Security and Safety

View full text Add to dashboard Cite

In this paper, we propose Device Authentication Code (DAC), a novel method for authenticating IoT devices with wireless interface, by exploiting their radio frequency (RF) signatures. The proposed DAC is based on RF fingerprinting, an information-theoretic method, feature learning, and the discriminatory power of deep learning. Specifically, an autoencoder is used to automatically extract features from the RF traces and the reconstruction error is used as the DAC, and this DAC is unique to each individual device. Then Kolmogorov-Smirnov (K-S) test is used to match the distribution of the reconstruction error generated by the receiver and the DAC in the received message, and the result will determine whether the device of interest is an intruder. We validate this concept on two experimentally collected RF traces from six ZigBee devices and five universal software defined radio peripheral devices, respectively. The traces span a range of Signal-to-Noise Ratio by varying locations, mobility of the devices, channel interference, and noise to ensure robustness of the model. Experimental results demonstrate that DAC is able to prevent device impersonation by extracting salient features that are unique to each wireless device of interest and can be used to identify radio frequency devices. Furthermore, the proposed method does not need the RF traces of the intruder during model training to be able to identify devices not seen during training, which makes it practical.

show abstract

Detection of Video Anomalies Using Convolutional Autoencoders and One-Class Support Vector Machines

Cited by 26 publications

References 12 publications

DeepFall: Non-Invasive Fall Detection with Deep Spatio-Temporal Convolutional Autoencoders

DeepFall: Non-Invasive Fall Detection with Deep Spatio-Temporal Convolutional Autoencoders

Anomalous Example Detection in Deep Learning: A Survey

Device Authentication Codes based on RF Fingerprinting using Deep Learning

Contact Info

Product

Resources

About