We describe an approach to generating animated sequences of images drawn using stroke-based rendering. Individual strokes are generated as a history of movements of a class of purely reactive computational agents known as Braitenberg Vehicles embedded on a digital canvas. The entire animation represents an aggregate of all the movements of the agents across an entire run. We provide end users with a software tool to produce such animations and allow users to set the sensory capabilities of agents on a low level basis. The agents perform feature detection based on such capabilities. The end users thus interact with the agents with a variety of input images to discover a range of renderable outcomes. Artists inform us that the animations that result from our output provide an engaging visual experience.
Unfortunately, ACM prohibits us from displaying non-influential references for this paper.
To see the full reference list, please visit http://dl.acm.org/citation.cfm?id=2001864.