MIL Based Anomaly Detection in Surveillance Videos Using I3D and Clip Feature Extractor
Shahriar Iqbal Shahriar, Al Fattah Siddique, Monideepa Kundu · 2024
This work presents an efficient anomaly detection method for surveillance videos using I3D and CLIP feature extractors. By leveraging a deep Multiple Instance Learning (MIL) framework with weakly labeled data, our approach avoids the need for time-consuming temporal annotations. The model is trained on the UCF Crime dataset, containing various real-world anomalies. Our method significantly outperforms previous approaches, achieving an AUC of 84.07 with I3D feature extractor, demonstrating its effectiveness in identifying anomalous events. This provides a practical and scalable solution for real-world surveillance applications, requiring minimal annotation effort.