Score: 0

Recognition of Abnormal Events in Surveillance Videos using Weakly Supervised Dual-Encoder Models

Published: November 17, 2025 | arXiv ID: 2511.13276v1

By: Noam Tsfaty , Avishai Weizman , Liav Cohen and more

Potential Business Impact:

Finds unusual things in videos automatically.

Business Areas:
Image Recognition Data and Analytics, Software

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.

Country of Origin
🇮🇱 Israel

Page Count
3 pages

Category
Computer Science:
CV and Pattern Recognition