System and method for object extraction

Abstract

A system and method for capturing images from various angles of an object that is resting on an orientation indication background, using a camera of a mobile device and visible light. The images are processed to construct a three-dimensional image model and a representation of the three-dimensional image model are displayed on an electronic display screen and manipulated or embedded into a video game.

We claim: 
     
         1 . A computer-implemented method comprising:
 (A) capturing, with a two-dimensional (2-D) image capture device, multiple distinct 2-D images of an arbitrary object, said object at a fixed position on a region having a known background pattern thereon, said multiple 2-D images corresponding to multiple orientations or positions of said 2-D image capture device relative to said object;   (B) in at least some of said multiple 2-D images, separating the object from the known background pattern to form multiple silhouettes corresponding to said object;   (C) obtaining a three-dimensional (3D) image model of said object, said obtaining using at least some of the multiple silhouettes; and then   (D) embedding a digital representation of said three-dimensional image model into an electronic display of said object.   
     
     
         2 . The method of  claim 1  wherein the capturing in (A) uses visible light. 
     
     
         3 . The method of  claim 1  wherein the known background pattern is visible. 
     
     
         4 . The method of  claim 1  wherein, in (B), said separating uses spectral features of the object or the background pattern. 
     
     
         5 . The method of  claim 1  wherein, in (B), said separating uses spatial features of the object or the background pattern. 
     
     
         6 . The method of  claim 1  wherein, in (B), said separating uses color features of the object or the background pattern. 
     
     
         7 . The method of  claim 1  wherein the object made from material selected from: plastic, paper, clay, and metal. 
     
     
         8 . The method of  claim 1  wherein the region having a known background pattern thereon is a mat. 
     
     
         9 . The method of  claim 1  wherein the region having a known background pattern thereon comprises a retroreflective material. 
     
     
         10 . The method of  claim 9  wherein the 2-D image capture device uses a lighting source to perform said capturing in (A). 
     
     
         11 . The method of  claim 10  wherein said lighting source is an LED. 
     
     
         12 . The method of  claim 1  wherein the known background pattern comprises at least two colors. 
     
     
         13 . The method of  claim 12  wherein the known background pattern comprises a grid. 
     
     
         14 . The method of  claim 13  wherein the known background pattern comprises two color grid. 
     
     
         15 . The method of  claim 1  wherein the known background pattern comprises a visible orientation indicator. 
     
     
         16 . The method of  claim 1  wherein the region having a known background pattern thereon comprises known features including one or more of: at least one specific color, and at least one specific spatial pattern. 
     
     
         17 . The method of  claim 1  wherein the known background pattern has known spatial features. 
     
     
         18 . The method of  claim 1  wherein the region having a known background pattern thereon comprises a color calibration palette. 
     
     
         19 . The method of  claim 1  further comprising:
 color calibrating said three-dimensional image model using said known background pattern. 
 
     
     
         20 . The method of  claim 1  further comprising:
 calibrating said image capture device during said capturing. 
 
     
     
         21 . The method of  claim 1  wherein obtaining the 3D image model of the object in (C) comprises: performing voxel carving on said at least some of said multiple silhouettes to obtain the 3D image model of said object. 
     
     
         22 . The method of  claim 21  wherein said voxel carving is performed in (C) while 2-D images of the object are being captured in (A). 
     
     
         23 . The method of  claim 22  wherein said voxel carving is performed in (C) in real time while 2-D images of the object are being captured. 
     
     
         24 . The method of  claim 1  where said separating in (B) comprises removing the known background pattern from said at least some of said multiple 2-D images. 
     
     
         25 . The method of  claim 1  wherein said separating in (B) is performed simultaneously or in real time with the capturing of the images in (A) 
     
     
         26 . The method of  claim 1 , wherein said electronic display is for a video game, and wherein the digital representation can be manipulated by a user in said video game. 
     
     
         27 . The method of  claim 1  wherein the 2-D image capture device is in a device selected from: a notebook, a netbook, a personal digital assistant (PDA), a mobile phone, a tablet, a pad, and a computer. 
     
     
         28 . The method of  claim 1  wherein said multiple 2-D images are captured in uncontrolled lighting conditions. 
     
     
         29 . The method of  claim 1  wherein the known background pattern is printed on a substrate. 
     
     
         30 . The method of  claim 29  wherein the substrate is paper or cardboard. 
     
     
         31 . The method of  claim 1  wherein the region having the known background pattern is on a rotatable surface and wherein, during said capturing in (A), said object is rotated with said rotatable surface to obtain multiple views of said object on said region.