What are geometry images and how can I create them ? It's a little weird ... I need to create geometry images from mesh models and vice versa.
1 Answers
Geometry images are a way of encoding shapes. They were introduced in this paper.
To quote from the introduction:
[A geometry image] captures geometry as a simple n × n array of [x, y,z] values. Other surface attributes, such as normals and colors, are stored as additional square images, sharing the same domain as the geometry. Because the geometry and attributes share the same parametrization, the parametrization itself is implicit — “texture coordinates” are absent. Moreover, this parametrization fully utilizes the texture domain (with no wasted space). Geometry images can be encoded using traditional image compression algorithm, such as wavelet based coders. Also, geometry images are ideally suited for hardware rendering. They may be transmitted to the graphics pipeline in a compressed form just like texture images. And, they eliminate expensive pointer-based structures such as indexed vertex lists.

- 18,863
- 8
- 57
- 82
-
Thanks .. I saw it but cannot understand ... What is this "parametrization"? Can anyone explain more ? – Rostamiani Apr 27 '12 at 12:52
-
@Rostamiani, it is the same as texture coordinates, which you can Google. – brainjam Apr 27 '12 at 15:48