Visualization and Animation Technology （Polygon Rendering）

Polygonal Mesh Rendering

The Polygonal Mesh Rendering or Polygonal Rendering is a viewing technique that allows for rendering 2D views of 3D models made by polygons.

The Polygonal Mesh Rendering exploits the geometrical characteristics of Polygons of being flat and identifiable from 3 vertices (顶点) and 1 normal vector (法向量) only.

The Polygonal Mesh Rendering is alternative to Ray-Tracing.

Advantages:

fast and allowing for hardware acceleration.

Disadvantages:

less realism (lack of shadows, reflections and refractions)

The curve surfaces are approximated because they need many polygons to describe.

Main Steps of Polygonal Rendering (The Graphic Pipeline)

1. Projective Transformation (投影变换)
   Establish the type of Projection: Perspective Projections and Parallel Projections.

2. Culling (剔除视锥外的objects)
   Identify the objects that are in the viewing frustum and therefore the only objects that will be rendered.

3. Scan Conversion (Projection & Display)
   The Scan Conversion maps polygon vertices to pixels with pixel depth and pixel colour. (Depth Buffer & Colour Buffer)

    	The depth of each polygon vertex is stored in the depth buffer.
    	The colour buffer stores the colour of each pixel in  the resulting 2D view.
   

4. Visible Surface Detection (Z-Buffer)
   Solve visibility issues through the use of Z-Buffer algorithm.
   1. Store a depth (Z) value for each pixel
   2. Store with Z set to a high (farthest distance from eye) value for each pixel
   3. As each polygon is rendered, update each pixel value and Z value if the new Z value is smaller (closer) than the old one.

The first two steps deal with geometry of the projected objects.
The last two steps deal with shading of the projected objects.

Z-Buffer algorithm

Advantage:

It correctly estimates visibility regardless of the order the polygons are considered.

Disadvantage:

It needs to take into account all polygons, including those that may be entirely or partially hidden.