Adaptive perspective ray casting

Authors:

Abstract
Ii'e present a method to accurately and eficientlyperform per- spective volumek.c ray casting of unjform regular datasets, called E~on&ntial-Begion (ER) Perspective. Unlike previous m&thodswhich undersample, oversimple, or approximate the data, our method near un#ormly samples the data through- out the viewing volume. In addition, it gajns algorithmic ad- vantagesfiom a regular sampling pattern and cache-coherent rad access, making itan algon-thm well suited for imple- mentation on hardware architectu~s for volume rendering. TT'equali~ the algorithm by its $ltering characteristics and demonstrate its effectiveness by wntrasting its antialiasing ~~aiity and timing with other perspective ray casting meth- ods. CR Categories: 1.3.3 (Computer Graphics): Pic- ture/Image Generation-D~ky dgorit~ 1.3.7 (Com- puter Graphics): Thr~DimensioA Graphics and W ~sm; 1.4.2 ~mage Processing And Computer Vision): Enhancement-Ffit*~