Skip to content

← ArchivePaper2024

Reconstructing Translucent Thin Objects from Photos

Xi Deng, Lifan Wu, Bruce Walter, Ravi Ramamoorthi, Eugene d'Eon, Steve Marschner, Andrea Weidlich

SIGGRAPH Asia6 citesCFX

This paper tackles joint reconstruction of shape and appearance for thin translucent objects such as leaves and paper from real photographs, where light transmission complicates capture.

Abstract

This paper tackles joint reconstruction of shape and appearance for thin translucent objects such as leaves and paper from real photographs, where light transmission complicates capture. It presents an affordable and fast acquisition pipeline that simultaneously recovers spatially varying reflectance and transmission through a two-phase optimization. The method reproduces the layered, translucent look of thin natural materials including feathers under varied lighting.

How to read this

Category
Method: shape-and-appearance reconstruction (inverse rendering) for thin translucent objects
Contributions
  • An affordable, fast acquisition pipeline for joint shape and appearance of thin translucent objects from real photos
  • A two-phase optimization that recovers spatially varying reflectance and transmission together
  • Reproduction of the layered translucent look of thin natural materials such as leaves, paper, and feathers under varied lighting
Context
An inverse-rendering capture method for thin translucent materials, related generally to spatially varying BRDF/BTDF acquisition and differentiable-rendering reconstruction (no specific prior works listed).
Correctness
Demonstrated on real photographs of thin natural materials where transmission complicates capture; results target thin objects specifically, so applicability to thick or strongly multiple-scattering media should not be assumed.
Clarity
Moderately technical; a first pass conveys the capture setup and goal, a second pass is needed for the two-phase optimization details.
How to read it
First pass for the acquisition setup and what reflectance-plus-transmission jointly buys you; deeper pass on the two-phase optimization if you intend to reproduce the pipeline.

Related work

Keywords

This page summarises the entry and links to its original source. The archive never hosts or redistributes the publication itself.Show it in the full archive list →