Gradient-based steering for vision-based crowd simulation algorithms

Additional Material Gradient-based steering for vision-based crowd simulation algorithms

Learning Crowd Behavior

We present here our ongoing work on learning crowd behavior. The steering behavior of crowds from various video sources is tracked and databases of examples are generated. These examples contain various stimuli (metrics) that could affect the persons behavior. These databases are used to learn rules for crowd steering in an agent based framework using regression algorithms and more specifically...

Learning Crowd Steering Behaviors from Examples

Crowd steering algorithms generally use empirically selected stimuli to determine when and how an agent should react. These stimuli consist of information from the environment that potentially influence behavior such as an agent’s visual perception, its neighboring agents state and actions, locations of nearby obstacles, etc. The various different approaches, rule-based, example-based or other,...

A Clustering Based Approach for Realistic and Efficient Data-Driven Crowd Simulation

In this paper, we present a data-driven approach to generate realistic steering behaviors for virtual crowds in crowd simulation. We take advantage of both rule-based models and data-driven models by applying the interaction patterns discovered from crowd videos. Unlike existing example-based models in which current states are matched to states extracting from crowd videos directly, our approac...

Pattern Based Motion for Crowd Simulation

We present a pattern-based approach for simulating the steering behavior of pedestrians, which aims to imitate the way that real pedestrians perceive spatial-temporal information and make steering decisions in daily-life situations. Novel representations of spatial-temporal patterns are proposed that allow modelers to intuitively and naturally specify some prototypical patterns for various stee...