Anomaly detection in surveillance videos using transformer based attention model

Abstract

Surveillance footage can catch a wide range of realistic anomalies. This research suggests using a weakly supervised strategy to avoid annotating anomalous segments in training videos, which is time consuming. In this approach only video level labels are used to obtain frame level anomaly scores. Weakly supervised video anomaly detection (WSVAD) suffers from the wrong identification of abnormal and normal instances during the training process. Therefore it is important to extract better quality features from the available videos. With this motivation, the present paper uses better quality transformer-based features named Videoswin Features followed by the attention layer based on dilated convolution and self attention to capture long and short range dependencies in temporal domain. This gives us a better understanding of available videos. The proposed framework is validated on real-world dataset i.e. ShanghaiTech Campus dataset which results in competitive performance than current stateof-the-art methods. The model and the code are available at https://github.com/kapildeshpande/Anomaly-Detection-in-Surve illance-Videos.

SwinAnomaly: Real-Time Video Anomaly Detection Using Video Swin Transformer and SORT

“…Transformers are more effective at simulating temporal dynamics than more conventional approaches like ConvL-STM and can minimize processing time. Deshpande et al [28] used a three-stage model consisting of a pre-trained videoswin model, an attention layer, and lastly an RFTM model for anomaly detection. TransUNet is a hybrid of U-Net with a transformer encoder, as suggested by Chen et al [29].…”

SwinAnomaly: Real-Time Video Anomaly Detection Using Video Swin Transformer and SORT

“…Transformers are more effective at simulating temporal dynamics than more conventional approaches like ConvLSTM and can minimize processing time. Deshpande et al [28] used a three-stage model consisting of a pre-trained videoswin model, an attention layer, and lastly an RFTM model for anomaly detection. TransUNet is a hybrid of U-Net with a transformer encoder, as suggested by Chen et al [29].…”