Image-based reconstruction of spatial appearance and geometric detail

Abstract

Real-world objects are usually composed of a number of different materials that often show subtle changes even within a single material. Photorealistic rendering of such objects requires accurate measurements of the reflection properties of each material, as well as the spatially varying effects. We present an image-based measuring method that robustly detects the different materials of real objects and fits an average bidirectional reflectance distribution function (BRDF) to each of them. In order to model local changes as well, we project the measured data for each surface point into a basis formed by the recovered BRDFs leading to a truly spatially varying BRDF representation. Real-world objects often also have fine geometric detail that is not represented in an acquired mesh. To increase the detail, we derive normal maps even for non-Lambertian surfaces using our measured BRDFs. A high quality model of a real object can be generated with relatively little input data. The generated model allows for rendering under arbitrary viewing and lighting conditions and realistically reproduces the appearance of the original object.

Intrinsic Textures for Relightable Free-Viewpoint Video

“…Lensch et al [22] introduce a robust method for the extraction of time-varying BRDF given a coarse geometric model of a real-world object. They propose extraction of surface normals as well as albedo in fitting the BRDF to give the illusion of high-frequency geometry.…”

Scene Reconstruction and Visualization From Community Photo Collections

“…There are many techniques for acquiring reflectance in the lab, using specialized equipment and controlled lighting including [13], [38], [46], and [80]. However, it is not possible to take objects like the Statue of Liberty or the Notre Dame Cathedral into the lab.…”

Building Illumination Coherent 3D Models of Large-Scale Outdoor Scenes

“…Without any knowledge about the illumination, however, this problem can become intractable. In laboratory conditions, the position of the light sources can be measured; with such measurements (Debevec et al 2000;Lensch et al 2003) show how to obtain a reflectance function once the light positions are known. To extend this idea to outdoor scenes, Yu and Malik (1998) measure the illumination of the scene by acquiring photographs of the sky and the surrounding environments.…”

“…This technique falls into the category of inverse-rendering techniques, since we are measuring scene properties from images and objects of known geometry. While measuring surface reflectance of an object of known geometry under controlled and calibrated illumination has proved to produce very good results as shown by Debevec et al (2000) and Lensch et al (2003), working with unknown illumination is yet an open problem. Typically, to handle unknown illumination it is assumed that the material properties of the object are homogeneous over the whole surface (Ikeuchi and Sato 1991;Ramamoorthi and Hanrahan 2001a).…”