Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Introducing Anisotropic Fields for Enhanced Diversity in Crowd Simulation

Yihao Li, Junyu Liu, Xiaoyu Guan, Hanming Hou, Tianyu Huang

TL;DR

This work proposes incorporating anisotropic fields (AFs) as a fundamental structure for depicting the uncertainty in crowd movement, and indicates that by incorporating AFs, crowd simulation systems can achieve a much higher similarity to real-world crowd systems.

Abstract

Large crowds exhibit intricate behaviors and significant emergent properties, yet existing crowd simulation systems often lack behavioral diversity, resulting in homogeneous simulation outcomes. To address this limitation, we propose incorporating anisotropic fields (AFs) as a fundamental structure for depicting the uncertainty in crowd movement. By leveraging AFs, our method can rapidly generate crowd simulations with intricate behavioral patterns that better reflect the inherent complexity of real crowds. The AFs are generated either through intuitive sketching or extracted from real crowd videos, enabling flexible and efficient crowd simulation systems. We demonstrate the effectiveness of our approach through several representative scenarios, showcasing a significant improvement in behavioral diversity compared to classical methods. Our findings indicate that by incorporating AFs, crowd simulation systems can achieve a much higher similarity to real-world crowd systems. Our code is publicly available at https://github.com/tomblack2014/AF\_Generation.

Introducing Anisotropic Fields for Enhanced Diversity in Crowd Simulation

TL;DR

This work proposes incorporating anisotropic fields (AFs) as a fundamental structure for depicting the uncertainty in crowd movement, and indicates that by incorporating AFs, crowd simulation systems can achieve a much higher similarity to real-world crowd systems.

Abstract

Large crowds exhibit intricate behaviors and significant emergent properties, yet existing crowd simulation systems often lack behavioral diversity, resulting in homogeneous simulation outcomes. To address this limitation, we propose incorporating anisotropic fields (AFs) as a fundamental structure for depicting the uncertainty in crowd movement. By leveraging AFs, our method can rapidly generate crowd simulations with intricate behavioral patterns that better reflect the inherent complexity of real crowds. The AFs are generated either through intuitive sketching or extracted from real crowd videos, enabling flexible and efficient crowd simulation systems. We demonstrate the effectiveness of our approach through several representative scenarios, showcasing a significant improvement in behavioral diversity compared to classical methods. Our findings indicate that by incorporating AFs, crowd simulation systems can achieve a much higher similarity to real-world crowd systems. Our code is publicly available at https://github.com/tomblack2014/AF\_Generation.
Paper Structure (21 sections, 9 equations, 12 figures, 3 tables, 1 algorithm)

This paper contains 21 sections, 9 equations, 12 figures, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Field-based Crowd Simulation
  4. Data-driven Crowd Simulation
  5. Sketch-based Control of Crowds
  6. Construction of Crowd Simulation Systems and Parameter Tuning
  7. Method
  8. Anisotropic Layer
  9. Definition of AFs
  10. Behavior Inertia of AFs
  11. Sketch-based Generation of AFs
  12. Data-driven AF Generation
  13. Evaluation
  14. Quantification of AFs Based on Information Entropy
  15. Comparison of Macroscopic Similarity in Crowd Systems
  16. ...and 6 more sections

Figures (12)

  • Figure 1: Potential impact of anisotropic fields on agent motion in a simulated scenario. (a): simulated scene. (b): illustration of an agent's behavioral diversity via AFs via two simulation iterations. Only one behavior occurs in the actual simulation; the result is uncertain.
  • Figure 2: Three-layer crowd simulation system after anisotropic layer addition. (a): overview. (b): details of anisotropic layer.
  • Figure 3: Difference between navigation fields and AFs. (a): In traditional navigation fields, agents execute deterministic motion based on the specified navigation directions. (b): In AFs, agents exhibit uncertainty in their movement.
  • Figure 4: (a) Graphical AF example, (b) Probability function. Agents tend to move toward directions II or IV in this AF region.
  • Figure 5: Behavioral inertia mechanism.
  • ...and 7 more figures