Exploring Convolutional Recurrent architectures for anomaly detection in videos: a comparative study

Ravi, Ambareesh; Karray, Fakhri

doi:10.1007/s44163-021-00004-2

Cited by 4 publications

(5 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These methods have demonstrated their effectiveness in capturing temporal patterns and detecting video anomalies. Some notable examples include Long Short-Term Memory (LSTM) [31,32], Gated Recurrent Unit (GRU) [33,34], and Convolutional Recurrent Neural Network (CRNN) [35].…”

Section: B Recurrent Neural Network (Rnns)mentioning

confidence: 99%

Investigating of Deep Learning-based Approaches for Anomaly Detection in IoT Surveillance Systems

HUANG,

CAI,

SUN

2023

IJACSA

View full text Add to dashboard Cite

Anomaly detection plays a crucial role in ensuring the security and integrity of Internet of Things (IoT) surveillance systems. Nowadays, deep learning methods have gained significant popularity in anomaly detection because of their ability to learn and extract intricate features from complex data automatically. However, despite the advancements in deep learning-based anomaly detection, several limitations and research gaps exist. These include the need for improving the interpretability of deep learning models, addressing the challenges of limited training data, handling concept drift in evolving IoT environments, and achieving real-time performance. It is crucial to conduct a comprehensive review of existing deep learning methods to address these limitations as well as identify the most accurate and effective approaches for anomaly detection in IoT surveillance systems. This review paper presents an extensive analysis of existing deep learning methods by collecting results and performance evaluations from various studies. The collected results enable the identification and comparison of the most accurate deep-learning methods for anomaly detection. Finally, the findings of this review will contribute to the development of more efficient and reliable anomaly detection techniques for enhancing the security and effectiveness of IoT surveillance systems.

show abstract

Section: B Recurrent Neural Network (Rnns)mentioning

confidence: 99%

Investigating of Deep Learning-based Approaches for Anomaly Detection in IoT Surveillance Systems

HUANG,

CAI,

SUN

2023

IJACSA

View full text Add to dashboard Cite

show abstract

“…A Time Distributed Layer (TDL) [23], shown as the large rectangle in Figure 1, was selected to compare a set of M sequential images and learn, in this way, the effect of transient changes (e.g. dynamic objects) to the predicted output.…”

Section: B Deep Convolutional Neural Networkmentioning

confidence: 99%

“…The LSTM layer is introduced to learn the temporal dependence among observations, e.g. image frames in chronological order [23]. A flattening or pooling operation after the TDL is introduced since the requirement is to have only one dimension per output from the Time Distributed wrapping to insert them to the LSTM layer.…”

Section: B Deep Convolutional Neural Networkmentioning

confidence: 99%

DNN-based anomaly prediction for the uncertainty in visual SLAM

Bosdelekidis

Johansen

Sokolova

2022

2022 17th International Conference on Control, Automation, Robotics and Vision (ICARCV)

View full text Add to dashboard Cite

The method described in this paper proposes a supervised Deep Neural Network (DNN) approach for the prediction of anomalies in camera-based navigation. The method is inspired by the unsolved issues of Integrity Monitors (IMs) when some of the sensor measurement covariances are unknown or inconsistent. Especially, the focus is on predicting when the estimation error distribution would require fatter tails to include outliers. The developed method takes into account single-frame image features as well as transient changes in the error. In the best of our knowledge, this is the first work that predicts anomalies in the error covariance of SLAM estimates and associates them with low-level image features. Finally, the prediction method can be used with other sensors as well, allowing the future development of navigation algorithm-and sensor-agnostic safety monitoring frameworks.

show abstract

“…In the next article, the authors Ambareesh Ravi, Fakhri Karray present a perspective, titled "Exploring Convolutional Recurrent Architectures for Anomaly Detection in Videos: a comparative study," the authors explore a variety of Convolutional Recurrent architectures and the influence of hyper-parameters on their performance for the task of anomaly detection [11].…”

Section: The Contents Of the Inaugural Issuementioning

confidence: 99%

The golden age of Artificial Intelligence

Kaynak

2021

Discov Artif Intell

View full text Add to dashboard Cite

The turing testAlso known as the imitation game, is a test to determine whether a machine can exhibit intelligent behaviour. It has been discussed extensively in AI Community as a controversial topic, and it appears that it will continue to be controversial. Even today, there is no consensus on the answer to the question Turing posed; "can machines think [3]?" The reference here is to a video clip, illustrating in black and white, the state of AI in the 1960s, in which legendary AI pioneers like Jerome Wiesner, Oliver Selfridge, and Claude Shannon speak. It is amazing to listen to the hyped expectations that are, 60 years later, still to come! Let there be AI! (Dartmouth Summer School 1956)The term Artificial Intelligence was coined in 1955 by John McCarthy and his colleagues Marvin Minsky (Harvard University), Nathaniel Rochester (IBM), and Claude Shannon (Bell Telephone Laboratories) in the proposal for the funding of "2-month, a 10-man study of Artificial Intelligence. " The workshop took place in the following year, in 1956, which is generally considered as the official birthdate of the new field. In their proposal, the proposers made very ambitious statements such as "(i) the study is to proceed based on the conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it, (ii) an attempt will be made to find how to make machines use language, form abstractions and concepts, solve kinds of problems now reserved for humans, and improve themselves. " LISPIn the 1960s, the most popular programming language used in AI research was LISP. McCarthy developed the basic ideas during 1956-1958, and it quickly became a common language for AI programming. The main reason for its popularity was that learning could be incorporated in LISP as self-modifying programs. The popularity of LISP was so high that special computers, so-called "LISP machines, " that could run LISP programs efficiently and effectively swarmed the market. The most notable machine was PDP-10 of Digital Equipment Corporation (DEC). Even I used one in my early years of research!

show abstract

Exploring Convolutional Recurrent architectures for anomaly detection in videos: a comparative study

Cited by 4 publications

References 32 publications

Investigating of Deep Learning-based Approaches for Anomaly Detection in IoT Surveillance Systems

Investigating of Deep Learning-based Approaches for Anomaly Detection in IoT Surveillance Systems

DNN-based anomaly prediction for the uncertainty in visual SLAM

The golden age of Artificial Intelligence

Contact Info

Product

Resources

About