Batch feature standardization network with triplet loss for weakly-supervised video anomaly detection

作者：

Highlights：

• A batch feature standardization module facilitates video anomaly detection.

• Research focus shifts from instance-level and bag-level relationships to batch-level associations.

• A novel strategy is used to introduce triplet loss in video anomaly detection.

• The method achieves comparable performance with the state-of-the-art approaches on several datasets.

摘要

•A batch feature standardization module facilitates video anomaly detection.•Research focus shifts from instance-level and bag-level relationships to batch-level associations.•A novel strategy is used to introduce triplet loss in video anomaly detection.•The method achieves comparable performance with the state-of-the-art approaches on several datasets.

论文关键词：Video anomaly detection,Batch feature standardization,Triplet loss,Feature processing

论文评审过程：Received 29 September 2021, Revised 12 January 2022, Accepted 26 January 2022, Available online 31 January 2022, Version of Record 15 February 2022.

论文官网地址：https://doi.org/10.1016/j.imavis.2022.104397