Parallax mapping

Parallax mapping with self-shading

Parallax mapping (also known as offset mapping or virtual displacement mapping ) is an improvement on bump mapping or normal mapping techniques used on textures in 3D image synthesis applications and in computer games.

For the end user ( game designer , player) this means that the representation of relief-like surfaces such as stone walls has more apparent depth and thus greater realism when using textures , without requiring the computational effort of displacement mapping .

Parallax mapping was published in 2001 by Tomomichi Kaneko et al. introduced.

Parallax mapping is generated by shifting texture coordinates at the point of the displayed polygon by a function of the viewing angle relative to the tangential plane (defined by the surface normal ) and the value of the height field at this point. At angles of view that are flatter relative to the tangential plane, the texture coordinates are shifted more and thus create the illusion of depth through parallax effects when changing the viewpoint.

Parallax mapping was introduced by Kaneko as a one-step process that ignores occlusion. Subsequent improvements to the algorithm, which added iterative methods, account for occlusion and the accurate representation of silhouettes.

Steep parallax mapping

Steep parallax mapping summarizes a class of algorithms that track lines of sight and test them against altitude fields. The underlying idea is to trace a ray that intersects the volume of a height field and find its intersection with the height field. This closest intersection determines which part of the height field is really visible. Relief mapping and parallax occlusion mapping are other well-known names for these techniques.

Interval mapping improves upon the binary search commonly performed in relief mapping by creating a line between points known to be inside and outside and intersecting them with a ray instead of using the binary search center as usual.