Acquisition and Modeling of Material Appearance Using a Portable, Low Cost, Device

Material appearance acquisition allows researchers to capture the optical properties of surfaces and use them in different tasks such as material analysis, digital twins reproduction, 3D configurators, augmented and virtual reality, etc. Precise acquisition of such properties requires complex and expensive hardware. In this paper, we aim to answer the following research challenge: Can we design an accurate enough but low-cost and portable device for material appearance acquisition? We present the rationale behind the design of our device using consumer-grade hardware components. Ultimately, our device costs EUR 80 and can acquire surface patches of size 5 × 5 cm with a 40 pix/mm resolution. Our device exploits a traditional RGB camera to capture a surface using 24 different images, each photographed using different lighting conditions. The different lighting conditions are generated by exploiting the LED rings included in our device; specifically, each of the 24 images is acquired by turning on one individual LED at time. We also illustrate the custom processing pipelines developed to support capturing and generating the material data in terms of albedo, normal, and roughness maps. The accuracy of the acquisition process is comprehensively evaluated both quantitatively and qualitatively. Results show that our low-cost device can faithfully acquire different materials. The usefulness of our device is further demonstrated by a textile virtual catalog application that we designed for rendering virtual fabrics on a mobile apparatus.