Image-space Caustics and Curvatures Xuan Yu Feng Li Jingyi Yu Department of Computer and Information Sciences University of Delaware Newark, DE 19716, USA fxuan,feng,
[email protected] Abstract Caustics are important visual phenomena, as well as challenging global illumination effects in computer graph- ics. Physically caustics can be interpreted from one of two perspectives: in terms of photons gathered on scene geom- etry, or in terms of a pair of caustic surfaces. These caustic surfaces are swept by the foci of light rays. In this paper, we develop a novel algorithm to approximate caustic sur- faces of sampled rays. Our approach locally parameterizes rays by their intersections with a pair of parallel planes. We show neighboring ray triplets are constrained to pass simul- taneously through two slits, which rule the caustic surfaces. We derive a ray characteristic equation to compute the two slits, and hence, the caustic surfaces. Using the characteris- tic equation, we develop a GPU-based algorithm to render the caustics. Our approach produces sharp and clear caus- tics using much fewer ray samples than the photon mapping method and it also maintains high spatial and temporal co- herency. Finally, we present a normal-ray surface repre- Figure 1. We use our caustic-surface-based sentation that locally parameterizes the normals about a algorithm to render the refraction caustics surface point as rays. Computing the normal ray caustic cast by a crystal bunny of 69473 triangles. On surfaces leads to a novel real-time discrete shape operator. an NVidia GeForce7800, our method renders at 115 fps at an image resolution of 512x512.