Shadow volume reconstruction from depth maps

Current graphics hardware can be used to generate shadows using either the shadow volume or shadow map techniques. However, the shadow volume techniqu e requires access to a representation of the scene as a polygonal model, an d handling the near plane clip correctly and efficiently is difficult; conv ersely, accurate shadow maps require high-precision texture map data repres entations, but these are not widely supported. We present a hybrid of the shadow map and shadow volume approaches which do es not have these difficulties and leverages high-performance polygon rende ring. The scene is rendered from the point of view of the light source and a sampled depth map is recovered. Edge detection and a template-based recon struction technique are used to generate a global shadow volume boundary su rface, after which the pixels in shadow can be marked using only a one-bit stencil buffer and a single-pass rendering of the shadow volume boundary po lygons. The simple form of our template-based reconstruction scheme simplif ies capping the shadow volume after the near plane clip.