A Survey of Anomalous Behavior Detection Techniques in Video Surveillance

Anomalous behavior detection in video surveillance is a central research problem in the field of intelligent security. With the large-scale deployment of surveillance cameras in public spaces, the need for automatic identification of anomalous events---such as fighting, falling, and unattended objects---from massive video streams has become increasingly urgent. This paper systematically reviews the mainstream technical approaches in this domain, encompassing traditional methods based on optical flow and background modeling, deep learning methods based on convolutional neural networks (CNNs) and long short-term memory networks (LSTMs), unsupervised anomaly detection methods based on autoencoders and generative adversarial networks (GANs), and the recently emerged Transformer and graph neural network (GNN) approaches. The paper further surveys widely adopted benchmark datasets---including UCF-Crime and ShanghaiTech---along with evaluation metrics such as frame-level area under the curve (AUC). The latest advances in multimodal fusion and edge deployment are discussed, and key challenges including high annotation costs, weak cross-scene generalization, and insufficient model interpretability are analyzed. Finally, the paper provides an outlook on frontier directions including vision-language large models and few-shot learning, aiming to serve as a comprehensive technical reference for researchers in this field.

• Publication year

  2026

• Venue

  International Journal of Emerging Technologies and Advanced Applications

• Publication date

  2026-02-26

• Fields of study

  Not labeled

• Identifiers
  DOI 10.62677/ijetaa.2601146
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Dynamic Erasing Network With Adaptive Temporal Modeling for Weakly Supervised Video Anomaly Detection

  2025cited by this paper
• Video anomaly detection in 10 years: a survey and outlook

  2024cited by this paper
• MTC-Net: Multi-granular Temporal Cascade Learning for Weakly Supervised Video Anomaly Detection

  2024cited by this paper
• Video Anomaly Detection via self-supervised and spatio-temporal proxy tasks learning

  2024cited by this paper
• Distilling Aggregated Knowledge for Weakly-Supervised Video Anomaly Detection

  2024cited by this paper
• Deep Learning-Based Anomaly Detection in Video Surveillance: A Survey

  2023cited by this paper
• Feature Prediction Diffusion Model for Video Anomaly Detection

  2023cited by this paper
• Weakly-Supervised Video Anomaly Detection With Snippet Anomalous Attention

  2023cited by this paper
• VadCLIP: Adapting Vision-Language Models for Weakly Supervised Video Anomaly Detection

  2023cited by this paper
• Unveiling the performance of video anomaly detection models - A benchmark-based review

  2023cited by this paper
• MGFN: Magnitude-Contrastive Glance-and-Focus Network for Weakly-Supervised Video Anomaly Detection

  2022cited by this paper
• Normalizing Flows for Human Pose Anomaly Detection

  2022cited by this paper
• A Hybrid Video Anomaly Detection Framework via Memory-Augmented Flow Reconstruction and Flow-Guided Frame Prediction

  2021cited by this paper
• Weakly-supervised Video Anomaly Detection with Robust Temporal Feature Magnitude Learning

  2021cited by this paper
• Self-Supervised Predictive Convolutional Attentive Block for Anomaly Detection

  2021cited by this paper
• A Hierarchical Spatio-Temporal Graph Convolutional Neural Network for Anomaly Detection in Videos

  2021cited by this paper
• Normal graph: Spatial temporal graph convolutional networks based prediction network for skeleton based video anomaly detection

  2020cited by this paper
• Not only Look, but also Listen: Learning Multimodal Violence Detection under Weak Supervision

  2020cited by this paper
• A Survey of Single-Scene Video Anomaly Detection

  2020cited by this paper
• Learning Memory-Guided Normality for Anomaly Detection

  2020cited by this paper
• Localizing Anomalies From Weakly-Labeled Videos

  2020cited by this paper
• Intelligent video surveillance: a review through deep learning techniques for crowd analysis

  2019cited by this paper
• Memorizing Normality to Detect Anomaly: Memory-Augmented Deep Autoencoder for Unsupervised Anomaly Detection

  2019cited by this paper
• Real-World Anomaly Detection in Surveillance Videos

  2018cited by this paper
• Remembering history with convolutional LSTM for anomaly detection

  2017cited by this paper
• Future Frame Prediction for Anomaly Detection - A New Baseline

  2017cited by this paper
• Learning Temporal Regularity in Video Sequences

  2016cited by this paper
• Gaussian Process Regression-Based Video Anomaly Detection and Localization With Hierarchical Feature Representation

  2015cited by this paper
• Anomaly Detection and Localization in Crowded Scenes

  2014influential reference
• Sparse reconstruction cost for abnormal event detection

  2011cited by this paper

• No citing papers are available for this paper.