This paper presents a novel implicit representation of solid models. With
this representation, every solid model can be effectively presented by three
layered depth-normal images (LDNIs) that are perpendicular to three orthogonal
axes respectively. The layered depth-normal images for a solid model, whose
boundary is presented by a polygonal mesh, can be generated efficiently with
help of the graphics hardware accelerated sampling. Based on this implicit
representation - LDNIs, solid modeling operations including the Boolean
operations and the offsetting operation have been developed. A contouring
algorithm is also introduced in this paper to generate thin structure and sharp
feature preserved mesh surfaces from the layered depth-normal images.
Comparisons between LDNIs and other implicit representation of solid models are
given at the end of the paper to demonstrate the advantages of LDNIs.