This is the repository for the paper Divide and Conquer in Video Anomaly Detection: A Comprehensive Review and New Approach.
The experimental code in this paper is based on the code from two papers, STG-NF and Jigsaw. Following the instructions provided in the papers, the replication of the research can be accomplished.
We replicated these two papers on different datasets and attempted the divide-and-conquer approach. The results are as follows:
We achieved the highest AUC score on the ShanghaiTech dataset, as of August 20, 2023. Please refer to the Paper with Code for more details.
