Skip to content
Conference

ViTBN: A Vision Transformer with Batch Normalization for Crowd Behaviour and Anomaly Detection in Surveillance Videos

Aug 2026 · International Conference Computational Vision and Bio Inspired Computing · pp. 1218-1225 · 0 citations · 21 references

Abstract

Intelligent crowd behaviour analysis is critical for modern video surveillance systems to enhance public safety and enable timely anomaly detection. This study presents ViTBN, a vision transformer-based framework augmented with batch normalization to improve feature robustness and the stability of training. The proposed framework processes frames extracted from surveillance videos and employs a pretrained Vision Transformer to capture long-range spatial relationships within individual frames. The extracted representations are subsequently processed through feature aggregation, Batch normalization, and a classification head for normal anomalous discrimination. Experimental evaluation on the UCF-Crime benchmark demonstrates an accuracy of 91.29%, precision of 83.46%, recall of 87.23%, and F1-score of 85.33%, respectively, indicating reliable discrimination between anomalous and normal surveillance activities. Ablation studies confirmed that incorporating batch normalization improved optimization stability and feature generalization compared to the baseline ViT setup. The findings suggest that integrating normalization layers into transformer-based architectures strengthens their applicability in surveillance contexts, offering a practical and effective approach to intelligent public safety systems in real-world settings

View source

Similar papers

Sep 2026

Anomaly Detection in Surveillance Video: An Unsupervised Unity Encoder Approach

A unique encoder-decoder based anomaly detection framework that incorporates skip connections to preserve spatial features during reconstruction and Tanh activation in the bottleneck for a balanced latent space to improve sensitivity to fine anomalies is proposed.

Manisha Sharma, R. Purwar · 0 citations
Conference Sep 2026

Comparative Analysis of YOLOv8 and Faster R-CNN for Real-Time Visual Anomaly Detection

Automated anomaly detection in video streams is essential for industrial process monitoring, security surveillance, and quality control systems. This paper presents a systematic comparative analysis of two state-of-the-art deep learning object detectors, YOLOv8 and Faster R-CNN, applied to the task of visual anomaly de...

S. Pchelintseva, S. Shiryaeva, I. Vakhrusheva · 0 citations
Open access Aug 2026

SmartFire Vision: An Attention-Pruned Hybrid Vision Transformer and Detection Transformer Framework for Accurate, Efficient, and Real-Time Fire and Smoke Detection in Smart City Video Surveillance

The SmartFire Vision framework provides a highly capable and computationally efficient means of fire detection, is particularly beneficial for CCTV-based smart city surveillance, and shows promising computational efficiency on desktop-class GPUs, though dedicated edge-hardware validation remains a direction for future...

Muhammad Azhar, Muhammad Arman, Asma Iqbal et al. · 0 citations
Aug 2026

Residual F-Lenet: an efficient hybrid deep learning model for abnormal event detection using surveillance videos

A ResidualF-Lenet for detecting abnormal events in surveillance videos is designed, which integrates Residual Network with Lenet and achieved improved levels of accuracy, True Positive Rate (TPR), and True Negative Rate (TNR) as 91.681, 90.740, and 92.802%, respectively.

P. S. Prakash, Rajalaxmi R. R. · 0 citations
Sep 2026

Focus and Supplement: Dual-Enhanced Vision Transformer for Multi-Class Anomaly Classification

Multi-class anomaly classification in industrial vision remains challenging due to noisy/incomplete anomaly representations and the unknown number of anomaly classes. To overcome this, we propose MACO, a novel multi-class anomaly classification framework that learns comprehensive representations and dynamically estimat...

Xu-Rui Li, En-Jie Xu, Chen-Zhou Li et al. · 0 citations
Sep 2026

Spatio-temporal video anomaly detection via CNN-ViT autoencoder and farneback optical flow

This research introduces a novel memory-augmented CNN-ConvViT autoencoder framework for unsupervised video anomaly detection and introduces a Temporal-Aware Prototype Memory Module (TAPMM) that explicitly learns normal spatio-temporal behavior patterns.

Vandana Pathak, Manoj Diwakar, Neeraj Kumar Pandey et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.