Method and apparatus for rendering a three-dimensional object from a two-dimensional image

摘要

Provided is a method and apparatus that may generate a three-dimensional (3D) object from a two-dimensional (2D) image, and render the generated 3D object. Light source information may be extracted from the 2D image using a characteristic of the 2D image, and property information associated with the 2D image may be extracted based on the light source information. In particular, specular information, scattering information, and the like may be stored in a database using a plurality of samples. Since specular information or scattering information similar to an input image may be retrieved from the data base, property information may be quickly extracted.

主权项

1. A method of rendering a three-dimensional (3D) object, comprising:
extracting light source information from a single two dimensional (2D) image; extracting property information associated with the single 2D image from a property information database based on the light source information, wherein the extracted property information comprises information of a generic 3D object shape similar in shape to an object within the single 2D image; generating the 3D object from the single 2D image; and rendering the 3D object based on the light source information and the property information using at least one processor.