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 Videos

Value-directed human behavior analysis from video using partially observable Markov decision processes.

Jesse Hoey1, James J Little

  • 1School of Computing, University of Dundee, Dundee, Scotland. jessehoey@computing.dundee.ac.uk

IEEE Transactions on Pattern Analysis and Machine Intelligence
|May 15, 2007
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

Affective connotations according to LLMs: implications for meaning measurement and cultural bias.

Cognition & emotion·2025
Same author

The Efficacy of Semantics-Preserving Transformations in Self-Supervised Learning for Medical Ultrasound.

Bioengineering (Basel, Switzerland)·2025
Same author

A survey of the impact of self-supervised pretraining for diagnostic tasks in medical X-ray, CT, MRI, and ultrasound.

BMC medical imaging·2024
Same author

User Perception of Smart Home Surveillance Among Adults Aged 50 Years and Older: Scoping Review.

JMIR mHealth and uHealth·2024
Same author

Disruption of Social Orders in Societal Transitions as Affective Control of Uncertainty.

The American behavioral scientist·2023
Same author

How Cultural Meanings of Occupations in the U.S. Changed During the Covid-19 Pandemic.

The American behavioral scientist·2023
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 method to learn human behavior models from video, enabling agents to understand actions and outcomes for utility maximization without needing expert labels.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Robotics

Background:

  • Understanding human behavior from video is crucial for human-agent interaction.
  • Current methods often rely on labeled data, limiting adaptability.

Purpose of the Study:

  • To develop a method for learning decision-theoretic models of human behaviors from video data.
  • To enable agents to infer the utility of observed actions and contexts.

Main Methods:

  • Utilized a partially observable Markov decision process (POMDP) integrated with dynamic Bayesian networks.
  • Employed a posteriori constrained optimization based on the expectation-maximization algorithm for parameter learning.
  • Created spatial and temporal abstractions from video observations for high-level decision-making.

Related Experiment Videos

Main Results:

  • The system automatically discovers relevant behavior classes and their importance for utility optimization.
  • Demonstrated effective learning of behavior-context-utility relationships.
  • Successfully applied the method to an imitation game, robotic control, and a card game.

Conclusions:

  • This approach obviates the need for labeled data, reducing expert bias.
  • Enables agents to learn and act upon the meaning of human behaviors in context.
  • Offers a robust framework for decision-theoretic behavior modeling from observational data.