Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: May 6, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

1.3K

Anomaly detection and localization in crowded scenes.

Weixin Li1, Vijay Mahadevan, Nuno Vasconcelos

  • 1University of California, San Diego, La Jolla.

IEEE Transactions on Pattern Analysis and Machine Intelligence
|November 16, 2013
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Additive-Free Contact-Electro-Catalysis/Vacuum Ultraviolet System for Rapid Mitigation of Antimicrobial-Resistance-Associated Contaminants in Water.

Research (Washington, D.C.)·2026
Same author

Compositional Analysis of Polymeric Proanthocyanidins from <i>Vitis amurensis</i> Rupr. (Vitaceae) Seeds After Catechin-Assisted Sulfitolytic Cleavage.

Foods (Basel, Switzerland)·2026
Same author

Identification of a saddle point (SP) for safe sacral segment channel planning via data analysis and rapid physical prototyping.

European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society·2026
Same author

Interfacial Covalent Bonds Between Fe-MoS<sub>2</sub> Quantum Dots and CoFe-MOF Triggering the Strain Effect for Efficient Overall Water Splitting.

Small (Weinheim an der Bergstrasse, Germany)·2026
Same author

Rapid Artificial Intelligence Autoplanning Rivals Manual Expert Planning for Cervical Brachytherapy.

Practical radiation oncology·2026
Same author

Effect of Acupuncture for Diminished Ovarian Reserve: A Randomized Sham-Controlled Trial.

International journal of women's health·2026
Same journal

TraGraph-GS: Trajectory Graph-based Gaussian Splatting for Arbitrary Large-Scale Scene Rendering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

SWIFT: A Small-World Interaction Framework for Flow-Aware Trajectory Prediction in Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Adaptive Hardness-Driven Dictionary Distillation for Incomplete Streaming View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Mixture of Global and Local Experts with Diffusion Transformer for Controllable Face Generation.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

This study introduces a novel joint detector for temporal and spatial anomalies in crowded scenes. The method achieves state-of-the-art results in detecting unusual behaviors using dynamic textures and saliency detection.

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Pattern Recognition

Background:

  • Detecting anomalous behaviors in crowded scenes is crucial for public safety and surveillance.
  • Existing methods often struggle to capture both spatial and temporal aspects of anomalies effectively.

Purpose of the Study:

  • To propose a joint detector for temporal and spatial anomalies in crowded scenes.
  • To develop a robust video representation that integrates appearance and dynamics for anomaly detection.

Main Methods:

  • Utilized a mixture of dynamic textures models for video representation.
  • Implemented a center-surround discriminant saliency detector for spatial anomaly scores.
  • Developed a normal behavior model from training data for temporal anomaly scores.

Related Experiment Videos

Last Updated: May 6, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

1.3K
  • Employed a conditional random field (CRF) for global consistency of anomaly judgments.

    • Main Results:

    • Defined multiscale spatial and temporal anomaly maps.
    • Achieved state-of-the-art anomaly detection performance on benchmark datasets.
    • Demonstrated effectiveness on a new dataset of crowded pedestrian walkways.

    Conclusions:

    • The proposed joint detector effectively identifies and localizes anomalous behaviors in crowded environments.
    • The integration of appearance, dynamics, and multiscale analysis enhances anomaly detection accuracy.
    • The method shows significant promise for real-world surveillance applications.