Skip to content

felipeek/dt-sgim

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

45 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

dt-sgim

Overview

Implementation of the DT-SGIM filter, published in Domain transform for spherical geometry images [1].

The DT-SGIM filter acts on polygonal meshes represented by Geometry Images obtained from spherical parametrization. As a result, the filter provides a smoothed mesh, while keeping its important curvatures.

example1

Compile

Linux

You need the dependencies: gcc, g++, make, GLFW, GLEW, libgl, libpng-dev and zlib-dev.

To compile simply run:

$ make

and the binary bin/gimmesh will be created.

OS X

You need the dependencies: gcc, g++, make, GLFW, GLEW and the OpenGL framework.

To compile simply run:

$ make

and the binary bin/gimmesh will be created.

Usage

To use this tool, you must provide a geometry image. They are represented as .gim files with the following format, in order:

  • Image Width: 4-bytes integer
  • Image Height: 4-bytes integer
  • Image Data: each pixel is represented as a 3-channel 4-byte float (RGB), totalizing 12 bytes per pixel. Total data size is (width * height * 3 (channels) * 4) bytes.

A set of example geometry images are available in the res/ folder.

To open the application, simply run:

$ ./bin/gimmesh -g <example.gim>

An iterative GUI will be opened and you will be able to filter the geometry image.

Wavefront Objects

It is also possible to transform wavefront objects to the .gim format using the application.

To do that, simply run:

./bin/gimmesh -o <example.obj>

You can also set the optional parameters:

-e <result.gim>	: specify the path of the geometry image that will be generated (default: ./export.gim)
-it <number>	: number of iterations for spherical parametrization algorithm (default: 500)
-s <number>	: size of geometry image (<n> x <n>) [must be an odd number] (default: 255)

Note: The parametrization used to transform meshes to geometry images still needs to be improved. Thus, the quality of the result will depend on the input mesh.

References