Continuous Locomotive Crowd Behavior Generation

2025-04-07

AI Summary

This paper addresses the challenge of generating continuous, realistic, and heterogeneous crowd motion trajectories—a limitation of existing methods. We propose the first end-to-end continuous crowd behavior generation framework. Methodologically, it integrates semantic segmentation-guided diffusion models for controllable spatial crowd layout, employs Markov chains to model individual behavioral diversity, and alternates between an emitter and a simulator to achieve long-horizon temporal modeling. Joint optimization of density maps and probabilistic trajectory maps ensures synergy between scene-level population dynamics and agent-level trajectory fidelity. Our contributions are threefold: (1) the first fully continuous crowd trajectory generation framework; (2) a user-controllable generation architecture accompanied by a novel benchmark evaluation protocol; and (3) significant improvements in trajectory realism and behavioral diversity across diverse geographic scenes. The code is publicly available.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Modeling and reproducing crowd behaviors are important in various domains including psychology, robotics, transport engineering and virtual environments. Conventional methods have focused on synthesizing momentary scenes, which have difficulty in replicating the continuous nature of real-world crowds. In this paper, we introduce a novel method for automatically generating continuous, realistic crowd trajectories with heterogeneous behaviors and interactions among individuals. We first design a crowd emitter model. To do this, we obtain spatial layouts from single input images, including a segmentation map, appearance map, population density map and population probability, prior to crowd generation. The emitter then continually places individuals on the timeline by assigning independent behavior characteristics such as agents' type, pace, and start/end positions using diffusion models. Next, our crowd simulator produces their long-term locomotions. To simulate diverse actions, it can augment their behaviors based on a Markov chain. As a result, our overall framework populates the scenes with heterogeneous crowd behaviors by alternating between the proposed emitter and simulator. Note that all the components in the proposed framework are user-controllable. Lastly, we propose a benchmark protocol to evaluate the realism and quality of the generated crowds in terms of the scene-level population dynamics and the individual-level trajectory accuracy. We demonstrate that our approach effectively models diverse crowd behavior patterns and generalizes well across different geographical environments. Code is publicly available at https://github.com/InhwanBae/CrowdES .

Problem

• Generates continuous, realistic crowd trajectories with heterogeneous behaviors.
• Models diverse crowd actions using a controllable emitter-simulator framework.
• Evaluates realism of generated crowds via scene and trajectory metrics.

Innovation

• Uses diffusion models for continuous crowd generation
• Employs Markov chain for diverse behavior simulation
• Integrates user-controllable emitter and simulator components