GitHub - google/practical-inverse-rendering-of-textured-and-translucent-appearance: SIGGRAPH 2025 "Practical Inverse Rendering Of Textured And Translucent Appearance" SIGGRAPH 2025 "Practical Inverse Rendering Of Textured And Translucent Appearance" - google/practical-inverse-rendering-of-textured-and-translucent-appearance

🚀 The source code for our #SIGGRAPH2025 paper "Practical Inverse Rendering of Textured and Translucent Appearance" is now available!
🔗 GitHub: github.com/google/pract...

For PDF, supplemental and high-quality videos, check out the project webpage at weiphil.github.io/portfolio/pr...! Joint work with Jérémy Riviere, Ruslan Guseinov, Stephan Garbin, Philipp Slusallek, @bbernd.bsky.social, Thabo Beeler and @deliovicini.bsky.social #SIGGRAPH2025 n/n

Finally, our methods enable high-quality facial appearance reconstruction from sparse captures. We fit parameters of standard production rendering models, producing realistic results without neural rendering. 5/n

Additionally, many real-world materials, e.g., skin, exhibit translucent appearance. We therefore also introduce an efficient differentiable rendering algorithm to recover textured, path-traced subsurface scattering parameters. 4/n

Our solution, Laplacian mipmapping, is a preconditioner combining differentiable mipmapping with a Laplacian pyramid. The benefits? Robust texture recovery while being computationally efficient, resolution-independent, and requiring minimal hyperparameter tuning. 3/n

Why is this challenging? Inverse rendering often suffers from scale mismatches, rendering noise and local minima. This prevents reliable recovery of textures, e.g., using standard Adam. In contrast, our approach achieves a more uniform and robust convergence. 2/n

Inverse rendering has become a standard tool for 3D reconstruction problems. However, recovering high-frequency appearance textures is challenging. In our SIGGRAPH 2025 paper, we propose several techniques to robustly reconstruct complex appearances (e.g., human skin). 1/n