Single-image Reflectance and Transmittance Estimation from Any Flatbed Scanner

Abstract

Flatbed scanners have emerged as promising devices for high-resolution, single-image material capture. However, existing approaches assume specific capture conditions that limit their scalability. This work introduces a method inspired by intrinsic image decomposition that removes shading and specularity, enabling material capture with standard flatbed scanners. It also extends single-image reflectance capture with opacity and transmittance estimation, key components of full material appearance for thin-layer materials such as fabrics.

Javier Rodríguez Vázquez

Principal Researcher

Principal Researcher in Artificial Intelligence working on generalizable autonomy, embodied AI, adaptive agents, planning, memory, and robot learning.