Learning to Rasterize Differentiably
Abstract
Differentiable rasterization changes the standard formulation of primitive rasterization -- by enabling gradient flow from a pixel to its underlying triangles -- using distribution functions in different stages of rendering, creating a "soft" version of the original rasterizer. However, choosing the optimal softening function that ensures the best performance and convergence to a desired goal requires trial and error. Previous work has analyzed and compared several combinations of softening. In this work, we take it a step further and, instead of making a combinatorial choice of softening operations, parameterize the continuous space of common softening operations. We study meta-learning tunable softness functions over a set of inverse rendering tasks (2D and 3D shape, pose and occlusion) so it generalizes to new and unseen differentiable rendering tasks with optimal softness.
Cite
@article{arxiv.2211.13333,
title = {Learning to Rasterize Differentiably},
author = {Chenghao Wu and Hamila Mailee and Zahra Montazeri and Tobias Ritschel},
journal= {arXiv preprint arXiv:2211.13333},
year = {2024}
}
Comments
Published at Computer Graphics Forum (EGSR 2024), code see https://github.com/Theo-Wu/MetaRas, project webpage see https://theo-wu.github.io/MetaRas/