The current version of our rendering system (7zip, 5mb) is based on Mitsuba Physically-based Renderer v0.4.4.
Please refer to Mitsuba's official document on how to compile and execute it under various operating systems.
Note: Our implementation (including a number of customized medium and volume data source classes) is
NOT well documented at the point.
We plan to release a cleaner version in the future.
All renderings in our ACM SIGGRAPH 2012 paper 'Structure-aware Synthesis for Predictive Woven Fabric Appearance' have been created using this exemplar database (7zip, 313mb) including 8 micro CT scans represented in Mitsuba's VOL format (see the document on GridVolume plugin for more details).
In addition, the data contains two example scenes (under /render) corresponding to Figure 13ab in the paper:
pattern4 | pattern5 |
Note: Rendering those scenes requires several giga-bytes of memory and a significant amount of computation.
Page maintained by Shuang Zhao Last update: March 2013 |