2.2 3D Scene Representation with Explicit Geometry – Geometry Based Representation

Geometry-based representations have been the fundamental form to represent 3D objects in the world for decades. Since the dawn of computer graphics, real world objects have been represented using geometric 3D surfaces with associated textures mapped onto them and then the surfaces are rasterized to generate the virtual view using graphics hardware. More sophisticated attributes such as appearance properties can be assigned and used to synthesize more realistic views as well. Point-based representations use a set of discrete samples on the surface to represent the geometry. The sophisticated attributes can be recorded with the surface sample. The sampling density can also be adjusted according to need. In addition, volumetric representation is developed as another branch of geometry-based representations. Volumetric techniques extend the concept of a set of 2D pixels representing a set of unit patches in an image to a set of 3D voxels representing a set of unit volumes of the 3D scene.

Fundamentally, the object can be created by three different methods: manual creation through a user interface such as Maya, results from physical simulation, and results from examination such as MRI or CT scan.

Animation describes the transformation of the objects including the rigid transformation and deformation. Geometry-based representations are generally easier to be manipulated than the image-based representation ...

Get 3D Visual Communications now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.