Image-based rendering and editing method and apparatus

A method of removing redundant data of a simple texture model, the method comprising:

• calculating normal vectors of pixels projected into a three-dimensional (3D) space by using death images among a plurality of simple texture images, the plurality of simple texture images rendering different aspects of a 3D model;
• calculating reliabilities of the pixels of each of the simple texture images by using the calculated normal vectors of the pixels;
• and
• comparing the reliabilities of pixels that originate from different simple texture images but render a same portion of the 3D model and removing the pixels having low reliabilities from the different simple texture images from which they respectively originate, wherein the calculation of the normal vectors of the pixels comprises:
• projecting the pixels of each of the simple texture images onto the 3D space based on the values of the pixels of each of the depth images and detecting the pixels adjacent to each of the pixels projected onto the 3D space;
• forming triangular planes using the pixels projected onto the 3D space and two adjacent pixels and calculating normal vectors of the triangular planes;
• and
• averaging the calculated normal vectors of the triangular planes.

A method of removing redundant data of a simple texture model, the method comprising:

• calculating normal vectors of pixels projected into a three-dimensional (3D) space by using depth images among a plurality of simple texture images, the plurality of simple texture images rendering different aspects of a 3D model;
• calculating reliabilities of the pixels of each of the simple texture images by using the calculated normal vectors of the pixels;
• and
• comparing the reliabilities of pixels that originate from different simple texture images but render a same portion of the 3D model and removing the pixels having low reliabilities from the different simple texture images from which they respectively originate,
• wherein in the calculation of the reliabilities of the pixels, the reliabilities of the pixels of each of the simple texture images are determined depending on inner products of the normal vectors of the pixels of each of the simple texture images and a vector perpendicular to a plane of vision of a camera used for creating the simple texture images.

A method of removing redundant data of a simple texture model, the method comprising:

• calculating normal vectors of pixels projected into a three-dimensional (3D) space by using depth images among a plurality of simple texture images, the plurality of simple texture images rendering different aspects of a 3D model;
• calculating reliabilities of the pixels of each of the simple texture images by using the calculated normal vectors of the pixels;
• and
• comparing the reliabilities of pixels that originate from different simple texture images but render a same portion of the 3D model and removing the pixels having low reliabilities from the different simple texture images from which they respectively originate,
• wherein in the comparing the reliabilities of the pixels, pixels of each of the depth images having low reliabilities are set as pixels of a background image.

A computer-readable recording medium, on which computer-readable program codes executing a method of removing redundant data of a simple texture model, the method comprising:

• calculating normal vectors of pixels projected into a three-dimensional (3D) space by using depth images among a plurality of simple texture images, the plurality of simple texture images rendering different aspects of a 3D model;
• calculating reliabilities of the pixels of each of the simple texture images by using the calculated normal vectors of the pixels;
• and
• comparing the reliabilities of pixels that originate from different simple texture images but render a same portion of the 3D model and removing the pixels having low reliabilities from the different simple texture images from which they respectively originate,
• wherein the calculation of the reliabilities of the pixels comprises considering pixels that have reliabilities that differ from each other by a predetermined critical value or less are as having a same reliability.

A computer-readable recording medium, on which computer-readable program codes executing a method of removing redundant data of a simple texture model, the method comprising:

• calculating normal vectors of pixels projected into a three-dimensional (3D) space by using depth images among a plurality of simple texture images, the plurality of simple texture images rendering different aspects of a 3D model;
• calculating reliabilities of the pixels of each of the simple texture images by using the calculated normal vectors of the pixels;
• and
• comparing the reliabilities of pixels that originate from different simple texture images but render a same portion of the 3D model and removing the pixels having low reliabilities from the different simple texture images from which they respectively originate,
• wherein the calculation of the normal vectors of the pixels comprises:
• projecting the pixels of each of the simple texture images onto the 3D space based on the values of the pixels of each of the depth images and detecting the pixels adjacent to each of the pixels projected onto the 3D space;
• forming triangular planes using the pixels projected onto the 3D space and two adjacent pixels and calculating normal vectors of the triangular planes;
• and
• averaging the calculated normal vectors of the triangular planes.

A computer-readable recording medium, on which computer-readable program codes executing a method of removing redundant data of a simple texture model, the method comprising:

• calculating normal vectors of pixels projected into a three-dimensional (3D) space by using depth images among a plurality of simple texture images, the plurality of simple texture images rendering different aspects of a 3D model;
• calculating reliabilities of the pixels of each of the simple texture images by using the calculated normal vectors of the pixels;
• and
• comparing the reliabilities of pixels that originate from different simple texture images but render a same portion of the 3D model and removing the pixels having low reliabilities from the different simple texture images from which they respectively originate,
• wherein in the calculation of the reliabilities of the pixels, the reliabilities of the pixels of each of the simple texture images are determined depending on inner products of the normal vectors of the pixels of each of the simple texture images and a vector perpendicular to a plane of vision of a camera used for creating the simple texture images.

A computer-readable recording medium, on which computer-readable program codes executing a method of removing redundant data of a simple texture model, the method comprising:

• calculating normal vectors of pixels projected into a three-dimensional (3D) space by using depth images among a plurality of simple texture images, the plurality of simple texture images rendering different aspects of a 3D model;
• calculating reliabilities of the pixels of each of the simple texture images by using the calculated normal vectors of the pixels;
• and
• comparing the reliabilities of pixels that originate from different simple texture images but render a same portion of the 3D model and removing the pixels having low reliabilities from the different simple texture images from which they respectively originate,
• wherein in the comparing the reliabilities of the pixels, pixels of each of the depth images having low reliabilities are set as pixels of a background image.

An apparatus removing redundant data of a simple texture model, the apparatus comprising:

• a normal vector calculating unit, which calculates normal vectors of pixels projected onto a three-dimensional (3D) space by using depth images among a plurality of simple texture images that render different aspects of a 3D model;
• a reliability calculating unit, which calculates reliabilities of the pixels of each of the simple texture images by using the calculated normal vectors obtained by the normal vector calculating unit;
• and
• a redundant data removing unit, which compares the reliabilities of the pixels originating from different simple texture images but rendering the same portion of the 3D model and removes the pixels having low reliabilities from the different simple texture images from which they respectively originate,
• wherein the reliability calculating unit considers pixels that have reliabilities that differ from each other by a predetermined critical value or less are as having a same reliability.

An apparatus removing redundant data of a simple texture model, the apparatus comprising:

• a normal vector calculating unit, which calculates normal vectors of pixels projected onto a three-dimensional (3D) space by using depth images among a plurality of simple texture images that render different aspects of a 3D model;
• a reliability calculating unit, which calculates reliabilities of the pixels of each of the simple texture images by using the calculated normal vectors obtained by the normal vector calculating unit;
• and
• a redundant data removing unit, which compares the reliabilities of the pixels originating from different simple texture images but rendering the same portion of the 3D model and removes the pixels having low reliabilities from the different simple texture images from which they respectively originate,
• wherein the normal vector calculating unit calculates the normal vectors of the pixels of each of the simple texture images by projecting the pixels of each of the simple texture images onto the 3D space based on the values of the pixels of each of the depth images and detecting the pixels adjacent to each of the pixels projected onto the 3D space, forming triangular planes using the pixels projected onto the 3D space and two adjacent pixels, calculating normal vectors of the triangular planes, and averaging the normal vectors of the triangular planes.

An apparatus removing redundant data of a simple texture model, the apparatus comprising:

• a normal vector calculating unit, which calculates normal vectors of pixels projected onto a three-dimensional (3D) space by using depth images among a plurality of simple texture images that render different aspects of a 3D model;
• a reliability calculating unit, which calculates reliabilities of the pixels of each of the simple texture images by using the calculated normal vectors obtained by the normal vector calculating unit;
• and
• a redundant data removing unit, which compares the reliabilities of the pixels originating from different simple texture images but rendering the same portion of the 3D model and removes the pixels having low reliabilities from the different simple texture images from which they respectively originate,
• wherein the reliability calculating unit determines the reliabilities of the pixels of each of the simple texture images depending on inner products of the normal vectors of the pixels of each of the simple texture images and a vector perpendicular to a plane of vision of a camera used for creating the simple texture images.

An apparatus removing redundant data of a simple texture model, the apparatus comprising:

• a normal vector calculating unit, which calculates normal vectors of pixels projected onto a three-dimensional (3D) space by using depth images among a plurality of simple texture images that render different aspects of a 3D model;
• a reliability calculating unit, which calculates reliabilities of the pixels of each of the simple texture images by using the calculated normal vectors obtained by the normal vector calculating unit;
• and
• a redundant data removing unit, which compares the reliabilities of the pixels originating from different simple texture images but rendering the same portion of the 3D model and removes the pixels having low reliabilities from the different simple texture images from which they respectively originate,
• wherein the redundant data removing unit sets the pixels of each of the depth images having low reliabilities as the pixels of a background image.