Volume Rendering Guided by Multiscale Medial Models
David T. Chen · 2010
We present a hybrid volume rendering method that uses object shape information provided by multiscale medial models to guide a splatting volume renderer. Such models can focus the rendering effort towards objects and their boundaries, thereby reducing the influence of noise in the volume data set and avoiding occlusion of the structures the user wants to inspect. We derive the most likely boundary point to be used for rendering given the image boundary and the boundary point specified by the multi-scale medial model. We show how a user can select regions of interest based on objects in order to omit occluding objects, and how object shape can be used to interpolate weak and noisy boundaries.