Determining intrinsic optical properties of 3D printing materials

Area of Science:

• Optical Engineering
• Materials Science
• Computational Imaging

Background:

• Accurate optical properties (spectral absorption, scattering, refractive index) are crucial for 3D printing applications like softproofing and material design.
• Existing measurement methods require specialized equipment and expertise, limiting accessibility.
• Designing materials that mimic optical characteristics, such as for dental restorations, necessitates precise optical data.

Purpose of the Study:

• To develop a novel, accessible method for determining the intrinsic optical properties of 3D printing materials.
• To leverage commercial reflectance/transmittance spectrophotometry for material characterization.
• To train machine learning models for inferring optical properties from measured data.

Main Methods:

• Utilized a commercial reflectance/transmittance spectrophotometer to gather material data.
• Modeled the spectrophotometer's light path using a Monte Carlo path tracer to simulate measurements.
• Trained machine learning models on simulated data to infer spectral absorption, scattering coefficients, and refractive index, incorporating smoothness as a regularization constraint.

Main Results:

• Successfully inferred intrinsic optical properties of 3D printing materials.
• Validated the method against accurate laboratory measurements for real printing materials.
• Developed and provided trained machine learning models for community use.

Conclusions:

• The proposed approach offers an accessible and accurate method for characterizing 3D printing materials' optical properties.
• This technique facilitates improved simulation of translucent prints and the design of advanced optical materials.
• The availability of trained machine learning models democratizes access to critical optical property data for researchers and developers.