Steering Behaviors for Autonomous Cameras

Abstract

International audienceThe automated computation of appropriate viewpoints in complex 3D scenes is a key problem in a number of computer graphics applications. In particular, crowd simulations create visually complex environments with many simultaneous events for which the computation of relevant viewpoints remains an open issue. In this paper, we propose a system which enables the conveyance of events occurring in complex crowd simulations. The system relies on Reynolds' model of steering behaviors to control and locally coordinate a collection of camera agents similar to a group of reporters. In our approach, camera agents are either in a scouting mode, searching for relevant events to convey, or in a tracking mode following one or more unfolding events. The key benefit, in addition to the simplicity of the steering rules, holds in the capacity of the system to adapt to the evolving complexity of crowd simulations by self-organizing the camera agents to track interesting events.Le placement automatique de caméra dans une scène 3D est un problème important en informatique graphique. En particulier, les simulations de foules produisent des scènes complexes pour lesquelles le choix du point de vue est un problème non résolu. Dans ce travail, nous présentons une approche permettant de déterminer le placement et le cadrage de plusieurs caméras évoluant dans une simulation de foule, de facon à montrer aux mieux les événements qui se déroulent dans la simulation. Outre sa simplicité, notre méthode présente l'avantage d'adapter automatiquement le comportement des caméras à la complexité de la scène

Narrative-driven camera control for cinematic replay of computer games

“…This occurs at the cost of loosing the exact composition of A and B in these regions, but improves the quality of the viewpoint. The proposition shows to be simpler than our previously proposed extension that relied on Bezier curves [Galvane et al 2013]. …”

“…Lino and Christie proposed a toric surface model that efficiently generates a range of viewpoints corresponding to the exact on-screen composition of two or three targets [Lino and Christie 2012]. The technique has been reused by [Galvane et al 2013] to create dedicated steering behaviors for autonomous cameras and has partly inspired the smooth camera animation system proposed in this paper. In this section, we give an overview of the method used to generate the cinematic replay from the extraction of the information to the generation of the camera rushes.…”

“…Drawing our inspiration from a model we previously developed [Galvane et al 2013], we propose a physically-based camera animation model that relies on forces directly driven by our extended toric surface or spherical surface and constrained by the 3D environment. The model considers the camera as an oriented particle, influenced by forces guiding both its position and orientation (i) towards the appropriate range of viewpoints satisfying a PSL specification (the positioning force) and (ii) avoiding collisions with obstacles in the environment (the containment force).…”

Menge: A Modular Framework for Simulating Crowd Movement

“…OpenSteer is a C++ application for exploring steering behaviors [57]. OpenSteer uses a plug-in architecture to introduce various scenarios such as, capture the flag, multiple pursuit, boids [24], waypoint following, soccer, and, curiously, pedestrians.…”

Generic Drone Control Platform for Autonomous Capture of Cinema Scenes

“…Offline solutions, such as [13] or [5] give interesting results but remain, by nature, unsuited to real-time path planning. Closer to our research topic, [6,7] introduced a reactive approach based on steering behaviors. Using the Prose Storyboard Language (PSL) -a shot description language introduced in [15] -as a way to define visual properties they propose a real-time solution that pushes the camera towards its optimal position while avoiding obstacles.…”