Skeletal Video Anomaly Detection using Deep Learning: Survey, Challenges and Future Directions

TL;DR

This survey reviews skeleton-based deep learning methods for video anomaly detection, emphasizing privacy protection and discussing challenges, taxonomy, and future research directions in the field.

Contribution

It provides a comprehensive taxonomy of skeleton-based anomaly detection algorithms and highlights privacy advantages over appearance-based methods.

Findings

Skeleton-based methods are privacy-preserving alternatives.

Current challenges include handling noise and complex motions.

Future research should focus on improving robustness and scalability.

Abstract

The existing methods for video anomaly detection mostly utilize videos containing identifiable facial and appearance-based features. The use of videos with identifiable faces raises privacy concerns, especially when used in a hospital or community-based setting. Appearance-based features can also be sensitive to pixel-based noise, straining the anomaly detection methods to model the changes in the background and making it difficult to focus on the actions of humans in the foreground. Structural information in the form of skeletons describing the human motion in the videos is privacy-protecting and can overcome some of the problems posed by appearance-based features. In this paper, we present a survey of privacy-protecting deep learning anomaly detection methods using skeletons extracted from videos. We present a novel taxonomy of algorithms based on the various learning approaches. We conclude that skeleton-based approaches for anomaly detection can be a plausible privacy-protecting alternative for video anomaly detection. Lastly, we identify major open research questions and provide guidelines to address them.