초록 ▼

A hand-manipulated prop is picked-up via a single video camera, and the camera image is analyzed to isolate the part of the image pertaining to the object for mapping the position and orientation of the object into a three-dimensional space, wherein the three-dimensional description of the object is

A hand-manipulated prop is picked-up via a single video camera, and the camera image is analyzed to isolate the part of the image pertaining to the object for mapping the position and orientation of the object into a three-dimensional space, wherein the three-dimensional description of the object is stored in memory and used for controlling action in a game program, such as rendering of a corresponding virtual object in a scene of a video display. Algorithms for deriving the three-dimensional descriptions for various props employ geometry processing, including area statistics, edge detection and/or color transition localization, to find the position and orientation of the prop from two-dimensional pixel data.

대표청구항 ▼

1. A method for controlling a game program comprising the steps of:manipulating a physical object within a field of view of a single video camera connected to a processing unit having a memory area and a display wherein said physical object comprises a cylinder having a solid color;supplying two-dim

1. A method for controlling a game program comprising the steps of:manipulating a physical object within a field of view of a single video camera connected to a processing unit having a memory area and a display wherein said physical object comprises a cylinder having a solid color;supplying two-dimensional pixel information picked up by said video camera to said processing unit;extracting at least one pixel group from said two-dimensional pixel information which corresponds to the physical object;determining a set of parameters from said at least one pixel group which define a two-dimensional geometric shape wherein said geometric shape defined by said extracted pixel group is a four-sided polygon;executing a predetermined algorithm on said set of parameters to map the definition of said two-dimensional geometric shape to a three-dimensional definition of a geometric volume, thereby providing a description of said geometric volume in three dimensions which corresponds to the position and orientation of the said manipulated physical object wherein said predetermined algorithm comprises the steps of:determining a length l of said pixel group;determining at least two widths w 1 and w 2 of said pixel group at different lengths along said pixel group;determining an inclination θ of the object in the X-Y plane of the image between the Y axis and said length l;determining an inclination Φ of the object in the Y-Z plane based on a ratio w 1 :w 2 of said two widths; anddetermining a depth Z of the object based on a phi-weighted length l φ of the pixel group or a phi-weighted number N φ of the total number of pixels making up the pixel group;storing said three-dimensional definition in said memory; andusing said description of said geometric volume to control action shown on said display. 2. A method for controlling a game program comprising the steps of:manipulating a physical object within a field of view of a single video camera connected to a processing unit having a memory area and a display wherein said physical object comprises a cylinder having a solid color;supplying two-dimensional pixel information picked up by said video camera to said processing unit:extracting at least one pixel group from said two-dimensional pixel information which corresponds to the physical object;determining a set of parameters from said at least one pixel group which define a two-dimensional geometric shape wherein the geometric shape defined by said extracted pixel group is a four-sided polygon;executing a predetermined algorithm on said set of parameters to map the definition of said two-dimensional geometric shape to a three-dimensional definition of a geometric volume, thereby providing a description of said geometric volume in three dimensions which corresponds to the position and orientation of the manipulated physical object wherein said predetermined algorithm comprises the steps of:determining a length l of said pixel group;determining at least two widths w 1 and w 2 of said pixel group at different lengths along said pixel group;determining an inclination θ of the object in the X-Y plane of the image between the Y axis and said length l;determining a depth Z of the object based on an average width w avg of the pixel group; anddetermining an inclination Φ of the object in the Y-Z plane based on the ratio l:w avg of said length l of the pixel group to the average width w avg ;storing said three-dimensional definition in said memory; andusing said description of said geometric volume to control action shown on said display. 3. A method for controlling a game program comprising the steps of:manipulating a physical object within a field of view of a single video camera connected to a processing unit having a memory area and a display wherein the physical object further comprises a sphere having one solid color mounted onto an end of a cylinder having another solid color;supplying two-dimensional pixel information picked up by said video camera to said processing unit;extracting at least one pixel group from said two-dimensional pixel information which corresponds to the physical object;determining a set of parameters from said at least one pixel group which define a two-dimensional geometric shape wherein the geometric shape defined by the extracted pixel group for said sphere is a circle, and the geometric shape defined by said extracted pixel group for said cylinder is a four-sided polygon;executing a predetermined algorithm on said set of parameters to map the definition of said two-dimensional geometric shape to a three-dimensional definition of a geometric volume, thereby providing a description of said geometric volume in three dimensions which corresponds to the position and orientation of the manipulated physical object wherein said predetermined algorithm comprises the steps of:determining a center point p s and a total pixel area of the extracted pixel group for said circle;determining another point p of a center line of the extracted pixel group for said four-sided polygon;determining a length l between said points p s and p;determining an inclination θ of the object in the X-Y plane of the image between the Y axis and said length l;determining a depth Z of the object based on a total number N of pixels in the pixel group for said circle; anddetermining an inclination Φ of the object in the Y-Z plane based on said length l between said points p s and p and said depth Z;storing said three-dimensional definition in said memory; andusing said description of said geometric volume to control action shown on said display. 4. A method for controlling a game program comprising the steps of:manipulating a physical object within a field of view of a single video camera connected to a processing unit having a memory area and a display wherein said physical object comprises a cylinder having one solid color, said cylinder having a plurality of circumferential stripes thereon, said stripes being of a different solid color than said one solid color;supplying two-dimensional pixel information picked up by said video camera to said processing unit;extracting at least one pixel group from said two-dimensional pixel information which corresponds to the physical object wherein said extracting step further comprises determining a plurality a color transitions between the cylinder and said stripes;determining a set of parameters from said at least one pixel group which define a two-dimensional geometric shape;executing a predetermined algorithm on said set of parameters to map the definition of said two-dimensional geometric shape to a three-dimensional definition of a geometric volume, thereby providing a description of said geometric volume in three dimensions which corresponds to the position and orientation of the manipulated physical object wherein said predetermined algorithm comprises the steps of:setting a threshold value D t defining a maximum separation in a two-dimensional chrominance color space;detecting color transitions based on a value D determined from combined chrominance signals Cr and Cb for both of the colors of said stripes and of said cylinder;determining that a color transition of sufficient magnitude has occurred when said value D exceeds said predetermined threshold;defining, by said color transitions of sufficient magnitude, respective lengths l 1 and l 2 between said stripes;determining an inclination Φ of the object in the Y-Z plane based on a ratio l 1 :l 2 of the lengths l 1 and l 2 between said stripes; anddetermining a depth Z of the object based on the sum of said respective lengths and said inclination Φ of the object in the Y-Z plane;storing said three-dimensional definition in said memory; andusing said description of said geometric volume to control action shown on said display. 5. A method for controlling a game program comprising the steps of:manipulatin g a physical object within a field of view of a single video camera connected to a processing unit having a memory area and a display wherein said physical object comprises a cylinder having one solid color, said cylinder having a helical stripe thereon, said helical stripe being of a different solid color than said one solid color;supplying two-dimensional pixel information picked up by said video camera to said processing unit;extracting at least one pixel group from said two-dimensional pixel information which corresponds to the physical object wherein said extracting step further comprises extracting at least one pixel group corresponding to said helical stripe;determining a set of parameters from said at least one pixel group which define a two-dimensional geometric shape;executing a predetermined algorithm on said set of parameters to map the definition of said two-dimensional geometric shape to a three-dimensional definition of a geometric volume, thereby providing a description of said geometric volume in three dimensions which corresponds to the position and orientation of the manipulated physical object;executing a predetermined algorithm for determining a degree of rotation of said object;storing said three-dimensional definition in said memory; andusing said description of said geometric volume to control action shown on said display. 6. The method according to claim 5, wherein said predetermined algorithm for determining a degree of rotation comprises the steps of:extracting a pixel group corresponding to said cylinder;defining a helix H from the pixel group corresponding to said helical stripe;determining a center line of the pixel group corresponding to said cylinder; anddetermining a height h between an end of the extracted pixel group corresponding to said cylinder and a point on said helix H where the center line intersects said helix H, wherein the degree of rotation of said cylinder is determined based on a ratio of said height h to a total length l of the pixel group corresponding to said cylinder.