CVNov 17, 2025
Recognition of Abnormal Events in Surveillance Videos using Weakly Supervised Dual-Encoder ModelsNoam Tsfaty, Avishai Weizman, Liav Cohen et al.
We address the challenge of detecting rare and diverse anomalies in surveillance videos using only video-level supervision. Our dual-backbone framework combines convolutional and transformer representations through top-k pooling, achieving 90.7% area under the curve (AUC) on the UCF-Crime dataset.