Saved in:
Bibliographic Details
Main Authors: Tsfaty, Noam, Weizman, Avishai, Cohen, Liav, Tshuva, Moshe, Aperstein, Yehudit
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.13276
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915622626000896
author Tsfaty, Noam
Weizman, Avishai
Cohen, Liav
Tshuva, Moshe
Aperstein, Yehudit
author_facet Tsfaty, Noam
Weizman, Avishai
Cohen, Liav
Tshuva, Moshe
Aperstein, Yehudit
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2511_13276
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Recognition of Abnormal Events in Surveillance Videos using Weakly Supervised Dual-Encoder Models
Tsfaty, Noam
Weizman, Avishai
Cohen, Liav
Tshuva, Moshe
Aperstein, Yehudit
Computer Vision and Pattern Recognition
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.
title Recognition of Abnormal Events in Surveillance Videos using Weakly Supervised Dual-Encoder Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.13276