(YIP-09) Improving Synthesis and Recognition of Crowded Scenes using Statistical Models of Group Behavior

Abstract

Virtual training environments are an important tool for training military personnel in a cost-effective way, and realistic crowd modeling and population simulation are crucial components for developing training scenarios set in urban environments. The key objectives of this project were to: (1) create computational models of human group behavior; (2) analyze group behavior in simulation environments; (3) synthetically generate realistic group behavior in software agents; and (4) simulate populations in large urban settings using agent-based models. In the future, these models will enable software developers to significantly improve the realism of agents in military simulations and the accuracy of behavior recognition in surveillance applications by better characterizing how social context affects action selection for humans in group settings.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 01, 2013
Accession Number
ADA578243

Entities

People

  • Gita Sukthankar

Organizations

  • University of Central Florida

Tags

Communities of Interest

  • Autonomy
  • Human Systems

DTIC Thesaurus Topics

  • Agent-Based Simulations
  • Air Force
  • Air Force Research Laboratories
  • Algorithms
  • Communities
  • Computational Science
  • Data Analysis
  • Generative Models
  • Machine Learning
  • Military Operations
  • Military Personnel
  • Models
  • Simulations
  • Social Media
  • Social Networks
  • Students
  • Training

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Computational Modeling and Simulation
  • Military Training and Readiness Simulation