A Microfacet-Based Model for Photometric Stereo with General Isotropic Reflectance.

Area of Science:

• Computer Vision
• Computer Graphics
• Material Science

Background:

• Photometric stereo relies on accurate surface reflectance models for shape reconstruction.
• Existing models often struggle to represent the full spectrum of surface appearances from diffuse to specular.
• A unified, physically-based model is needed for robust and versatile surface analysis.

Purpose of the Study:

• To develop a precise, stable, and invertible microfacet-based reflectance model for photometric stereo.
• To introduce a single variable for physically quantifying surface smoothness, enabling smooth transitions between reflectance types.
• To provide a compact, interpretable formulation with a fast solver for accurate surface shape and smoothness estimation.

Main Methods:

• Developed a microfacet-based reflectance model applicable to isotropic surfaces.
• Introduced a single smoothness variable (0-1) for versatile representation, transitioning from Lambertian to specular.
• Designed a fast, lightweight solver for inverse estimation of surface smoothness and shape.

Main Results:

• The proposed model accurately represents a wide range of surface reflectances, from diffuse to specular.
• The single smoothness variable provides a physically meaningful and versatile control.
• The fast solver achieves accurate estimations of surface smoothness and shape on synthesized and real-world data.

Conclusions:

• The developed reflectance model is state-of-the-art for photometric stereo applications.
• The model offers a unified and physically interpretable approach to surface reflectance.
• The efficient solver makes the model practical for real-time and off-the-shelf solutions.