Plateau-free Differentiable Path Tracing

by   Michael Fischer, et al.

Current differentiable renderers provide light transport gradients with respect to arbitrary scene parameters. However, the mere existence of these gradients does not guarantee useful update steps in an optimization. Instead, inverse rendering might not converge due to inherent plateaus, i.e., regions of zero gradient, in the objective function. We propose to alleviate this by convolving the high-dimensional rendering function that maps scene parameters to images with an additional kernel that blurs the parameter space. We describe two Monte Carlo estimators to compute plateau-free gradients efficiently, i.e., with low variance, and show that these translate into net-gains in optimization error and runtime performance. Our approach is a straightforward extension to both black-box and differentiable renderers and enables optimization of problems with intricate light transport, such as caustics or global illumination, that existing differentiable renderers do not converge on.


page 6

page 13


Differentiable Shadow Mapping for Efficient Inverse Graphics

We show how shadows can be efficiently generated in differentiable rende...

Differentiable Visual Computing

Derivatives of computer graphics, image processing, and deep learning al...

Inverse Path Tracing for Joint Material and Lighting Estimation

Modern computer vision algorithms have brought significant advancement t...

Inverse Global Illumination using a Neural Radiometric Prior

Inverse rendering methods that account for global illumination are becom...

Zero Grads Ever Given: Learning Local Surrogate Losses for Non-Differentiable Graphics

Gradient-based optimization is now ubiquitous across graphics, but unfor...

Accelerating Derivative-Free Optimization with Dimension Reduction and Hyperparameter Learning

We consider convex, black-box objective functions with additive or multi...

Learning to Rasterize Differentiable

Differentiable rasterization changes the common formulation of primitive...

Please sign up or login with your details

Forgot password? Click here to reset