Skip to content

shi-jian/mitsuba-shapenet

master
Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?
Code
This branch is 14 commits ahead, 3 commits behind mitsuba-renderer:master.

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
 
 
doc
 
 
 
 
src
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

mitsuba-shapenet

Most ShapeNet models are exported from SketchUp with double-faces. In original Mitsuba renderer without back-face culling, there are black pixels when intersecting the 'back side' face. There are also some categories, such as cars, are not double-sided. Some of them only have flipped faces, which makes backface culling not work.

This fork simply implements a 'shapenet' shape importer. It reads model and material from ShapeNet .obj files and assign them with proper 'twosided' BSDF.

For more details about mitsuba renderer, please visit its homepage at https://www.mitsuba-renderer.org/ .

And for the dependency of the project, please check https://www.mitsuba-renderer.org/repos/ .

On Windows, you can simply put the dependencies_windows repo under the project folder and rename it as mitsuba-shapenet\dependencies. It would automatically find required headers and libraries.

If you meet the problem of missing header such as mitsuba_precompiled_header.hpp, please turn off MTS_USE_PCH flag during CMake.

Currently it cannot handle per-vertex normal or smooth group in ShapeNet obj. You can use 'maxSmoothAngle' to add some smoothness rendering effect.

Samples

Rendering scripts and results can be found here. Images rendered with this ShapeNet importer(left) and Mitsuba OBJ importer(right):

alt text alt text

alt text alt text

About

Mitsuba for ShapeNet

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • C++ 91.4%
  • Objective-C 4.2%
  • C 2.5%
  • Python 0.9%
  • Objective-C++ 0.4%
  • GLSL 0.3%
  • Other 0.3%