Robust Unsupervised Video Anomaly Detection by Multi-Path Frame Prediction

  title={Robust Unsupervised Video Anomaly Detection by Multi-Path Frame Prediction},
  author={X. Wang and Zhengping Che and Ke Yang and Bo Jiang and Jian-Bo Tang and Jieping Ye and Jingyu Wang and Q. Qi},
  journal={IEEE transactions on neural networks and learning systems},
  • X. Wang, Zhengping Che, +5 authors Q. Qi
  • Published 5 November 2020
  • Medicine, Computer Science
  • IEEE transactions on neural networks and learning systems
Video anomaly detection is commonly used in many applications, such as security surveillance, and is very challenging. A majority of recent video anomaly detection approaches utilize deep reconstruction models, but their performance is often suboptimal because of insufficient reconstruction error differences between normal and abnormal video frames in practice. Meanwhile, frame prediction-based anomaly detection methods have shown promising performance. In this article, we propose a novel and… 

Figures and Tables from this paper

Situational Anomaly Detection in Multimedia Data under Concept Drift
Anomaly detection has been a very challenging and active area of research for decades, but most of the works detect predefined anomaly classes using static models, which has limited applicability for real-life surveillance where the data have concept drift.
Video Abnormal Event Detection by Learning to Complete Visual Cloze Tests
A novel approach named visual cloze completion (VCC), which performs VAD by learning to complete “visual cloze tests” (VCTs), and a new DNN architecture is designed as a model-level solution to utilize video semantics and temporal context.
Unsupervised Anomaly Detection for a Smart Autonomous Robotic Assistant Surgeon (SARAS) Using a Deep Residual Autoencoder
An unsupervised approach to anomaly detection for robotic MIS based on deep residual autoencoders to make the autoencoder learn the ‘normal’ distribution of the data and detect abnormal events deviating from this distribution by measuring a reconstruction error.


Spatio-Temporal AutoEncoder for Video Anomaly Detection
A novel model called Spatio-Temporal AutoEncoding (ST AutoEncoder or STAE), which utilizes deep neural networks to learn video representation automatically and extracts features from both spatial and temporal dimensions by performing 3-dimensional convolutions, which enhances the motion feature learning in videos.
Unsupervised Anomaly Detection and Localization Based on Deep Spatiotemporal Translation Network
A Deep Spatiotemporal Translation Network (DSTN), novel unsupervised anomaly detection and localization method based on Generative Adversarial Network (GAN) and Edge Wrapping that outperforms other state-of-the-art algorithms with respect to the frame-level evaluation, the pixel- level evaluation, and the time complexity for abnormal object detection and localized tasks.
Future Frame Prediction for Anomaly Detection - A New Baseline
This paper proposes to tackle the anomaly detection problem within a video prediction framework by introducing a motion (temporal) constraint in video prediction by enforcing the optical flow between predicted frames and ground truth frames to be consistent, and is the first work that introduces a temporal constraint into the video prediction task.
Real-World Anomaly Detection in Surveillance Videos
The experimental results show that the MIL method for anomaly detection achieves significant improvement on anomaly detection performance as compared to the state-of-the-art approaches, and the results of several recent deep learning baselines on anomalous activity recognition are provided.
Anomaly Detection in Video Sequence With Appearance-Motion Correspondence
A deep convolutional neural network that addresses this problem by learning a correspondence between common object appearances and their associated motions by designed as a combination of a reconstruction network and an image translation model that share the same encoder.
A Discriminative Framework for Anomaly Detection in Large Videos
An anomaly detection setting in which training sequences are unavailable and anomalies are scored independently of temporal ordering is addressed, which inspires a shift in approaches from classical density estimation to simple discriminative learning.
Traffic anomaly detection based on image descriptor in videos
A new traffic anomaly detection algorithm is introduced that firstly divides a traffic video into several video cubes in temporal domain, and each video cube is divided into video blocks in spatial domain.
Video anomaly detection and localization using hierarchical feature representation and Gaussian process regression
A hierarchical framework for detecting local and global anomalies via hierarchical feature representation and Gaussian process regression, which outperforms the main state-of-the-art methods on this topic and achieves at least 80% detection rates based on three challenging datasets.
Learning Deep Representations of Appearance and Motion for Anomalous Event Detection
This work proposes Appearance and Motion DeepNet (AMDN) which utilizes deep neural networks to automatically learn feature representations, and introduces a novel double fusion framework, combining both the benefits of traditional early fusion and late fusion strategies.
Abnormal Event Detection in Videos using Spatiotemporal Autoencoder
This work proposes a spatiotemporal architecture for anomaly detection in videos including crowded scenes that includes two main components, one for spatial feature representation, and one for learning the temporal evolution of the spatial features.