Crowdbrush

Abstract

Recent advances in computer graphics techniques and increasing power of graphics hardware made it possible to display and animate large crowds in real-time. Most of the research efforts have been directed towards improving rendering or behavior control; the question how to author crowd scenes in an efficient way is usually not addressed. We introduce a novel approach to create complex scenes involving thousands of animated individuals in a simple and intuitive way. By employing a brush metaphor, analogous to the tools used in image manipulation programs, we can distribute, modify and control crowd members in real-time with immediate visual feedback. We define concepts of operators and instance properties that allow to create and manage variety in populations of virtual humans. An efficient technique allowing to render up to several thousands of fully three-dimensional polygonal characters with keyframed animations at interactive framerates is presented. The potential of our approach is demonstrated by authoring a scenario of a virtual audience in a theater and a scenario of a pedestrian crowd in a city.

Interaction Fields: Intuitive Sketch‐based Steering Behaviors for Crowd Simulation

“…Various methods exist that try to give users more intuitive control over a crowd via sketch‐based techniques. Most of these provide control over global paths , by letting users draw curves for agents to follow [UdHT04; KSA∗09; OO09; MM17; BBEK20] or directional hints that are converted to a flow field [PWJ∗08; PvdBC∗11; KK∗14]. Some of these methods also propose sketch‐based control of other simulation parameters, such as the walking speed and the smoothness of paths [UdHT04; OO09].…”

“…Most of these provide control over global paths , by letting users draw curves for agents to follow [UdHT04; KSA∗09; OO09; MM17; BBEK20] or directional hints that are converted to a flow field [PWJ∗08; PvdBC∗11; KK∗14]. Some of these methods also propose sketch‐based control of other simulation parameters, such as the walking speed and the smoothness of paths [UdHT04; OO09]. Other work focuses more on changing the geometry of the environment [MMHR16; MM19; SGH20], on fitting a group into a formation [GD13; HSK14], or on giving users high‐level control over where agents go and which actions they perform [MR05; KSRF11; KBK16; MBA20].…”

A natural language interface with casual users for crowd animation

“…Crowd simulation techniques have been proposed, such as social force models, 1 power‐law, 2 models based on psychological factors, 3 radial‐view‐based method, 22 probability‐based method, 23 interaction in groups, 5 and environmental attributes 24 . Editing tools have been developed successfully 8,9,25 …”

Generation of small groups with rich behaviors from natural language interface

“…Crowd generation and simulation . Crowd can be created by painting with crowd attributes in a map 18 . A semantic virtual environment can support heterogeneous behaviors of crowds 19 .…”