Juan Raúl Padrón Griffe

About me
I am a researcher in computer graphics at the Graphics and Imaging Lab at the Universidad de Zaragoza, working at the intersection of physically-based rendering, material modeling, and geometry processing. My research centers on multi-scale materials, from biological tissues such as human skin, scales, and feathers, to intricate human-made structures like cosmetics and granular media. Recently, I have also explored particle dynamics for sampling in computer graphics, including our recent paper accepted to EGSR 2026. My work has been published at different computer graphics venues including the Eurographics Symposium on Rendering (EGSR), Pacific Graphics (PG), and SIGGRAPH. I am currently collaborating with Zahra Montazeri's group on the representation and rendering of textiles, with a submission under review and a second project underway.

I completed my PhD as a Marie Sklodowska-Curie Fellow of the EU Project PRIME, supervised by Prof. Adolfo Muñoz and Prof. Adrian Jarabo. Earlier, during my Master of Science in Informatics at the Technical University of Munich, I conducted research on 3D scanning and neural rendering for object and face relighting, advised by Prof. Justus Thies; this followed a Bachelor of Science in Computer Science at the Universidad Central de Venezuela, where my thesis explored procedural terrain generation and visualization.

Looking for Opportunities
I recently defended my Ph.D. dissertation, Modeling and Rendering of Multiscale Materials. I am currently seeking postdoctoral and faculty positions, as well as research scientist roles, where I can continue developing my research agenda in computer graphics and computer vision for the digital acquisition, representation, and simulation of virtual worlds. If you are building a research group, have a postdoctoral or faculty opening, or simply want to discuss a potential collaboration, please feel free to reach out!

Projects

Dynamical System for Blue Noise (EGSR 2026)

A unified particle dynamics framework for spectral noise point distribution synthesis, transforming arbitrary input distributions into blue, pink, and red noise patterns within a single simulation pipeline. We showcase applications of the framework on image stippling, object placement, and Monte Carlo rendering. The framework is implemented in Python (NumPy, SciPy's cKDTree, and Numba) and results visualized via Blender's bpy API and evaluated in Monte Carlo rendering through PBRT v4.

2 minute read

Procedural Multiscale Geometry (PG 2025)

A procedural framework for modeling multiscale volumetric materials using implicit surfaces and sphere tracing. It supports rich micro- and mesostructures, including microstructure reconstruction from image and distance values. The framework is implemented as procedural shaders in the NVIDIA OptiX ray tracing API. This project received a Best Paper (Honorable Mention) award.

2 minute read

Pennaceous Feathers Rendering (PG 2024)

Far-field appearance model that implicitly represents the light scattering from the main biological structures of a feather such as the shaft, barb and barbules accounting for the accurate masking and the medulla structure inside the barbs. The material was implemented as a BSDF material inside the popular physically-based renderer Mitsuba, while the geometry was represented by a 2D texture.

3 minute read

Foundation Cosmetics Rendering (EGSR 2024)

Multilayered appearance model inspired by the specific characteristics of cosmetics foundation components capable of stacking multiple cosmetic products on top of a digital human skin. The material was implemented as a BSDF layered material inside the popular physically-based renderer PBRT.

2 minute read

Snake Skin Rendering (CEIG 2023)

Multilayered appearance model based on the anatomy of the snake skin capable of reproducing the main appearance features on snake colors such as the highly specular iridescent colors and dark diffuse skin. The material was implemented as a BSDF layered material inside the popular physically-based renderer Mitsuba.

1 minute read

Face Relighting In The Wild

Given an arbitrary portrait image and a target lighting as inputs, the algorithm generates the relight version of the portrait image under the target lighting conditions.

2 minute read

Neural Relighting

Extension of the Deferred Neural Rendering pipeline to perform relighting tasks. This is a new paradigm for image synthesis that combines the traditional graphics pipeline with learnable Neural Textures.

1 minute read