Enhancing Weakly-Supervised Video Anomaly Detection With Temporal Constraints

Anomaly Detection has been a significant field in Machine Learning since it began gaining traction. In the context of Computer Vision, the increased interest is notorious as it enables the development of video processing models for different tasks without the need for a cumbersome effort with the an...

Full description

Saved in:
Bibliographic Details
Main Authors: Francisco Caetano, Pedro Carvalho, Christina Mastralexi, Jaime S. Cardoso
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10965669/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850170899383189504
author Francisco Caetano
Pedro Carvalho
Christina Mastralexi
Jaime S. Cardoso
author_facet Francisco Caetano
Pedro Carvalho
Christina Mastralexi
Jaime S. Cardoso
author_sort Francisco Caetano
collection DOAJ
description Anomaly Detection has been a significant field in Machine Learning since it began gaining traction. In the context of Computer Vision, the increased interest is notorious as it enables the development of video processing models for different tasks without the need for a cumbersome effort with the annotation of possible events, that may be under represented. From the predominant strategies, weakly and semi-supervised, the former has demonstrated potential to achieve a higher score in its analysis, adding to its flexibility. This work shows that using temporal ranking constraints for Multiple Instance Learning can increase the performance of these models, allowing the focus on the most informative instances. Moreover, the results suggest that altering the ranking process to include information about adjacent instances generates best-performing models.
format Article
id doaj-art-a8fcc806521e481086bfbbbdc2aaa5d9
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-a8fcc806521e481086bfbbbdc2aaa5d92025-08-20T02:20:23ZengIEEEIEEE Access2169-35362025-01-0113708827089410.1109/ACCESS.2025.356076710965669Enhancing Weakly-Supervised Video Anomaly Detection With Temporal ConstraintsFrancisco Caetano0Pedro Carvalho1https://orcid.org/0000-0003-4983-4316Christina Mastralexi2https://orcid.org/0009-0002-0992-8801Jaime S. Cardoso3https://orcid.org/0000-0002-3760-2473INESC TEC—Institute for Systems and Computer Engineering, Technology, and Science, Porto, PortugalINESC TEC—Institute for Systems and Computer Engineering, Technology, and Science, Porto, PortugalINESC TEC—Institute for Systems and Computer Engineering, Technology, and Science, Porto, PortugalINESC TEC—Institute for Systems and Computer Engineering, Technology, and Science, Porto, PortugalAnomaly Detection has been a significant field in Machine Learning since it began gaining traction. In the context of Computer Vision, the increased interest is notorious as it enables the development of video processing models for different tasks without the need for a cumbersome effort with the annotation of possible events, that may be under represented. From the predominant strategies, weakly and semi-supervised, the former has demonstrated potential to achieve a higher score in its analysis, adding to its flexibility. This work shows that using temporal ranking constraints for Multiple Instance Learning can increase the performance of these models, allowing the focus on the most informative instances. Moreover, the results suggest that altering the ranking process to include information about adjacent instances generates best-performing models.https://ieeexplore.ieee.org/document/10965669/Multiple instance learningtemporal constraintsvideo anomaly detectionweakly-supervised models
spellingShingle Francisco Caetano
Pedro Carvalho
Christina Mastralexi
Jaime S. Cardoso
Enhancing Weakly-Supervised Video Anomaly Detection With Temporal Constraints
IEEE Access
Multiple instance learning
temporal constraints
video anomaly detection
weakly-supervised models
title Enhancing Weakly-Supervised Video Anomaly Detection With Temporal Constraints
title_full Enhancing Weakly-Supervised Video Anomaly Detection With Temporal Constraints
title_fullStr Enhancing Weakly-Supervised Video Anomaly Detection With Temporal Constraints
title_full_unstemmed Enhancing Weakly-Supervised Video Anomaly Detection With Temporal Constraints
title_short Enhancing Weakly-Supervised Video Anomaly Detection With Temporal Constraints
title_sort enhancing weakly supervised video anomaly detection with temporal constraints
topic Multiple instance learning
temporal constraints
video anomaly detection
weakly-supervised models
url https://ieeexplore.ieee.org/document/10965669/
work_keys_str_mv AT franciscocaetano enhancingweaklysupervisedvideoanomalydetectionwithtemporalconstraints
AT pedrocarvalho enhancingweaklysupervisedvideoanomalydetectionwithtemporalconstraints
AT christinamastralexi enhancingweaklysupervisedvideoanomalydetectionwithtemporalconstraints
AT jaimescardoso enhancingweaklysupervisedvideoanomalydetectionwithtemporalconstraints