Abstract: Techniques for generating a fully-textured 3D model of a human face are disclosed. According to one aspect of the present invention, a fully automatic system is configured to generate a 3-D model from a sequence of images that are produced from a single camera when a user is moving his/her head steadily from one direction to another, for example, from the left side of the camera to the right side of the camera. The images are processed in a two-stage process. First, the camera pose information is estimated for each of the valid image frames. Second, a mesh model is generated with respect to the camera pose information, and subsequently is textured with reference to one or more of the image frames.

Abstract: Techniques and systems that provide interactions between a 3D representation of a selected pair of glasses and a fully-textured 3D face model are disclosed. According to one embodiment, an interactive platform is displayed to allow a user to select a pair of glasses and try the selected glasses on a user-provided 3D face model. The interactions provided in the platform include spatial adjustments of the glasses around the face model, various perspective views of the 3D face with the glasses on and other cosmetic alternations to the selected glasses. According to one application, when the user finishes the try-on process, the information about the glasses can be transmitted to a business that can subsequently produce a pair of customized glasses for the user.

Abstract: A 3D modeling system for automatically generating fully-textured 3D models of objects from a sequence of images taken around the objects is disclosed. There are several processes developed to facilitate the operation of the 3D modeling system by an ordinary skilled person. One of the processes is the automatic calibration of a camera using only a portion of a calibration disc to essentially provide a larger effective field of view of the camera. Another process is a space carving process that subdivides volumetric cells recursively to fit to a 3D object using a tree structure that encodes the entire process. Still another process is a 3D mesh model generation process that begins with the tree structure and generates self-constraint and interconnected triangles, in a sense that all triangles intersect with each other either not at all or at common boundary faces, to represent the shape of the 3D object.

Abstract: Techniques for generating a 3D region from a surrounding imagery are disclosed. The surrounding imagery comprising a sequence of side view images taken respectively at a known position around an object. The images are preprocessed to generate corresponding sequence of mask images. To accelerate the subsequent space carving process, each of the mask images is encoded uniquely using what is called herein a Maximal Area Encoding scheme. In the space carving process, volumetric cells or cubes representing the object are recursively subdivided to gradually fit the object by “carving” away those cubes that are not occupied by any of the object. All cubes are encoded and indexed using an octree structure.

Abstract: A 3D modeling system employing a commercially available structured light source is disclosed. The light source projecting a light stripe is swung manually across an object or a particular area of interest on the surface of the object. Under the illumination of the swinging light stripe, the object is imaged to produce a sequence of images. For each of the images, the position and orientation of the structured light source are independently and dynamically derived. From each of the images, a set of curvilinear points from the (deformed) projected line falling on the object or the area of interest are calculated in conjunction with a calibration model of the imager. The curvilinear points from all the images are merged to provide accurate and complete 3D data about the object or the area of interest.

Abstract: A mesh generator and process based on a volumetric representation is disclosed. According to one aspect of the present invention, a 3D region of an object is generated by a space carving process and represented by volumetric cells encoded in a tree structure, such as an octree tree. Given the 3D region, the mesh process generates a mesh of elements covering the entire 3D region and intersecting with each other either not at all or at common boundary faces.

Abstract: A method and apparatus for generating exportable patches are disclosed. The exportable patches representing a mesh model are respectively generated using a test-and-merge procedure. A patch is a collection of the polygons with the property that every polygon in the patch shares at least one edge with some other polygons in the same patch. In addition, all patches have the properties that the union of all the polygons contains all the polygons of the mesh, and that no two patches contain the same polygon. Exporting such patches in image files makes it possible for a user to alter or modify the texture mapping for a particular patch in a desirable way.

Abstract: A 3D modeling system employing a commercially available structured light source is disclosed. The light source projecting a light stripe is swung manually across an object or a particular area of interest on the surface of the object. Under the illumination of the swinging light stripe, the object is imaged to produce a sequence of images. For each of the images, the position and orientation of the structured light source are independently and dynamically derived. From each of the images, a set of curvilinear points from the (deformed) projected line falling on the object or the area of interest are calculated in conjunction with a calibration model of the imager. The curvilinear points from all the images are merged to provide accurate and complete 3D data about the object or the area of interest.