Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Porous Flow and Sediment Transport Simulation for Physically-Based Weathering

Theo Jonchier 1 Arthur Cavalier 1, 1 Thibault Tricard 2 Guillaume Gilet 3 Stephane Merillou 4 
2 MFX - Matter from Graphics
Inria Nancy - Grand Est, LORIA - ALGO - Department of Algorithms, Computation, Image and Geometry
Abstract : In the graphic community, rendering lifelike scenes remains an open challenge. Among the features required to reach photorealism, aging is the crucial detail that breaks the pristine aspect that is rarely observed in real life. Multiple approaches have been proposed, from example-based to physical simulation, to compute the aging of an object. In this context, we present a simulation framework that handles multiple weathering effects causing an object to alter over time. We identify the key phenomena (wetting, drying, erosion, deposition, and dissolution) and propose a base framework to address the aging process. To tackle this challenge, our method adapts a Smooth Particle Hydrodynamics model to represent external fluids (i.e., classical fluid simulation) and internal porous flow. Our framework handles the wetting, drying, and flow in the porous space in a unified approach. We extend pre-existing sediment transport method to allow the sediment transport outside and inside the porous space. With our method, sediments can be eroded from a solid surface, transported across solid objects through their porous space, and deposited to modify other objects’ properties across the scene.
Complete list of metadata
Contributor : Theo Jonchier Connect in order to contact the contributor
Submitted on : Wednesday, June 15, 2022 - 10:34:54 PM
Last modification on : Thursday, June 23, 2022 - 1:11:18 PM


  • HAL Id : hal-03696332, version 1


Theo Jonchier, Arthur Cavalier, Thibault Tricard, Guillaume Gilet, Stephane Merillou. Porous Flow and Sediment Transport Simulation for Physically-Based Weathering. 2022. ⟨hal-03696332⟩



Record views


Files downloads