Refer to Lecture09.md
.
...
Implicit: Points satisfy some specified relationship. Example:
Pros:
Compact description
Certain queries easy (inside object, distance to surface)
Good for ray-to-surface intersection
For simple shapes, exact description / no sampling error
Easy to handle changes in topology
Cons:
Difficult to model complex shapes
Examples:
Algebraic Surfaces
Constructive Solid Geometry: Combine implicit geometry via Boolean operations
Level Set Methods: Store a grid of values approximating a function. Surface is found where interpolated values equal zero
Distance Functions: Given minimum distance (could be signed distance) from anywhere to object
Blend a moving boundary
Blend any two distance functions
Fractals: Exhibit self-similarity, detail at all scales
Explicit: All points are given directly or via parameter mapping. Example:
Easy to sample
Hard to test inside/outside
Examples:*
Point Cloud
Easiest representation
Useful for large datasets
Often converted into polygon mesh
Difficult to draw in under-sampled regions
Polygon Meshes
Store vertices & polygons (often triangles or quads)
Easier to do processing / simulation, adaptive sampling
More complicated data structures
Perhaps the most common representation in graphics
Format Example:
The Wavefront Object File (.obj
) Format
Text file specifying vertices, normals ,texture coordinates and their connectivity (by specifying for example the triangular face, with normals and texture coordinates on three vertices)
Bezier Surfaces
Polygon Mesh
Subdivision Surfaces
NURBS
...