MTFL: Multi-Timescale Feature Learning for Weakly-Supervised Anomaly Detection in Surveillance Videos

Yiling Zhang, Erkut Akdag, Egor Bondarev, Peter H. N. De With
10/8/2024
cs.CV

No official code found. Best implementation: erktkdg/MTFL (11 stars, last commit 2026-03-04).

Abstract

Detection of anomaly events is relevant for public safety and requires a combination of fine-grained motion information and contextual events at variable time-scales. To this end, we propose a Multi-Timescale Feature Learning (MTFL) method to enhance the representation of anomaly features. Short, medium, and long temporal tubelets are employed to extract spatio-temporal video features using a Video Swin Transformer. Experimental results demonstrate that MTFL outperforms state-of-the-art methods on the UCF-Crime dataset, achieving an anomaly detection performance 89.78% AUC. Moreover, it performs complementary to SotA with 95.32% AUC on the ShanghaiTech and 84.57% AP on the XD-Violence dataset. Furthermore, we generate an extended dataset of the UCF-Crime for development and evaluation on a wider range of anomalies, namely Video Anomaly Detection Dataset (VADD), involving 2,591 videos in 18 classes with extensive coverage of realistic anomalies.

View on arXivView PDF

Code Implementations(1)

erktkdg/MTFLCommunity100%

MTFL: Multi-Timescale Feature Learning for Weakly-Supervised Anomaly Detection in Surveillance Videos

116CSS, HTMLJun 1, 20257 months ago

Similar Papers

ComplexVAD: Detecting Interaction Anomalies in Video

89% match2025

VADTree: Explainable Training-Free Video Anomaly Detection via Hierarchical Granularity-Aware Tree

88% match1 repo192025

Exploring the Magnitude-Shape Plot Framework for Anomaly Detection in Crowded Video Scenes

88% match2024

Learning Event Completeness for Weakly Supervised Video Anomaly Detection

88% match2026

Dynamic Distinction Learning: Adaptive Pseudo Anomalies for Video Anomaly Detection

87% match1 repo112024

Pistachio: Towards Synthetic, Balanced, and Long-Form Video Anomaly Benchmarks

87% match2025

Cite this paper

@article{zhang2024mtfl,
  title  = {MTFL: Multi-Timescale Feature Learning for Weakly-Supervised Anomaly Detection in Surveillance Videos},
  author = {Yiling Zhang and Erkut Akdag and Egor Bondarev and Peter H. N. De With},
  year   = {2024},
  eprint = {2410.05900},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2410.05900v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2410.05900.svg)](https://codeofpaper.com/paper/2410.05900)

Discussion