There is a newer version of SoftRasSoft Rasterizer (SoftRas) is a truly differentiable renderer framework with a novel formulation that views rendering as a differentiable aggregating process that fuses probabilistic contributions of all mesh triangles with respect to the rendered pixels. Thanks to such "soft" formulation, our framework is able to (1) directly render colorized mesh using differentiable functions and (2) back-propagate efficient supervision signals to mesh vertices and their attributes (color, normal, etc.) from various forms of image representations, including silhouette, shading and color images.
Accessing SoftRas f644279-fosscuda-2019b-Python-3.7.4
To load the module for SoftRas f644279-fosscuda-2019b-Python-3.7.4 please use this command on the BEAR systems (BlueBEAR, BEARCloud VMs, and CaStLeS VMs):
module load SoftRas/f644279-fosscuda-2019b-Python-3.7.4
BEAR Apps Version
For more information visit the SoftRas website.
This version of SoftRas has a direct dependency on: fosscuda/2019b Python/3.7.4-GCCcore-8.3.0 PyTorch/1.3.1-fosscuda-2019b-Python-3.7.4 scikit-image/0.16.2-fosscuda-2019b-Python-3.7.4 SciPy-bundle/2019.10-fosscuda-2019b-Python-3.7.4 torchvision/0.4.2-fosscuda-2019b-Python-3.7.4 tqdm/4.41.1-GCCcore-8.3.0
These versions of SoftRas are available on the BEAR systems (BlueBEAR, BEARCloud VMs, and CaStLeS VMs). These will be retained in accordance with our Applications Support and Retention Policy.
|Version||BEAR Apps Version|
Last modified on 4th February 2020