Method and apparatus for encoding a video signal of a contour of an object

Claim 1

A method for encoding a video signal of a current contour of an object based on a previous contour thereof, comprising the steps of: (a) determining a multiple number of first vertex points on the current contour; (b) mapping the first vertex points onto the previous contour to thereby provide second vertex points of said multiple number on the previous contour; (c) approximating the current contour based on the first vertex points to thereby provide a first approximation contour; (d) approximating the previous contour based on the second vertex points to thereby provide a second approximated contour; (e) obtaining a first set of approximation errors between the current contour and the first approximated contour and a second set of approximation errors between the previous contour and the second approximated contour; (f) calculating a difference between the first and the second sets of approximation errors; (g) encoding the difference to produce encoded data; and (h) providing an encoded video signal of the current contour, the encoded video signal including the encoded data and vertex information representing locations of the first vertex points.

Claim 2

A method for encoding a video signal of a current contour of an object based on a previous contour thereof, comprising the steps of: (a) determining a multiple number of first vertex points on the current contour; (b) mapping the first vertex points onto the previous contour to thereby provide second vertex points of said multiple number on the previous contour; (c) approximating the current contour based on the first vertex points to thereby provide a first approximation contour; (d) approximating the previous contour based on the second vertex points to thereby provide a second approximated contour; (e) obtaining a first set of approximation errors between the current contour and the first approximated contour and a second set of approximation errors between the previous contour and the second approximated contour; (f) calculating a difference between the first and the second sets of approximation errors; (g) encoding the difference to produce encoded data; and (h) providing an encoded video signal of the current contour, the encoded video signal including the encoded data and vertex information representing locations of the first vertex points.

Claim 8

A method for encoding a video signal of a current contour of an object based on a previous contour thereof, comprising the steps of: (a) determining a multiple number of first vertex points on the current contour; (b) approximating a current contour segment formed by two first vertex points with a first line segment connecting said two first vertex points, said two first vertex points being adjacent each other along the current contour; (c) mapping said two first vertex points onto the previous contour to thereby generate two corresponding second vertex points on the previous contour; (d) approximating a previous contour segment defined by said two second vertex points with a second line segment joining said two second vertex points; (e) setting N number of sample points on the first and the second line segments, respectively, N being a positive integer; (f) providing a first set of approximation errors measured with respect to the current contour segment at the sample points on the first line segment and a second set of approximation errors calculated with respect to the previous contour segment at the sample points on the second line segment; (g) generating a set of differences between the first and the second sets of approximation errors; and (h) encoding the set of differences to thereby provide encoded data.

Claim 9

A method for encoding a video signal of a current contour of an object based on a previous contour thereof, comprising the steps of: (a) determining a multiple number of first vertex points on the current contour; (b) approximating a current contour segment formed by two first vertex points with a first line segment connecting said two first vertex points, said two first vertex points being adjacent each other along the current contour; (c) mapping said two first vertex points onto the previous contour to thereby generate two corresponding second vertex points on the previous contour; (d) approximating a previous contour segment defined by said two second vertex points with a second line segment joining said two second vertex points; (e) setting N number of sample points on the first and the second line segments, respectively, N being a positive integer; (f) providing a first set of approximation errors measured with respect to the current contour segment at the sample points on the first line segment and a second set of approximation errors calculated with respect to the previous contour segment at the sample points on the second line segment; (g) generating a set of differences between the first and the second sets of approximation errors; and (h) encoding the set of differences to thereby provide encoded data.

Claim 14

An apparatus for use in a video signal encoder, for compressing a video signal of a current contour of an object with respect to a previous contour thereof, which comprises: means for determining a plurality of first vertex points on the current contour; means for providing, based on the first vertex points, second vertex points of said plurality on the previous contour wherein each second vertex point corresponds to one of the first vertex points; means for approximating the current and the previous contours by a first and a second sets of line segments, respectively, wherein each line segment of the first set is formed by joining two of the first vertex points positioned adjacent each other on the current contour and each line segment of the second set is formed by connecting two of the second vertex points disposed neighboring each other along the previous contour; means for generating a first set of errors between a current contour segment and a line segment of the first set which are defined by two adjacent first vertex points and a second set of errors between a previous contour segment and a line segment of the second set which are determined by two second vertex points corresponding to said two adjacent first vertex points; means for calculating a difference between the first and the second sets of errors; and means for encoding the difference.

Claim 15

An apparatus for use in a video signal encoder, for compressing a video signal of a current contour of an object with respect to a previous contour thereof, which comprises: means for determining a plurality of first vertex points on the current contour; means for providing, based on the first vertex points, second vertex points of said plurality on the previous contour wherein each second vertex point corresponds to one of the first vertex points; means for approximating the current and the previous contours by a first and a second sets of line segments, respectively, wherein each line segment of the first set is formed by joining two of the first vertex points positioned adjacent each other on the current contour and each line segment of the second set is formed by connecting two of the second vertex points disposed neighboring each other along the previous contour; means for generating a first set of errors between a current contour segment and a line segment of the first set which are defined by two adjacent first vertex points and a second set of errors between a previous contour segment and a line segment of the second set which are determined by two second vertex points corresponding to said two adjacent first vertex points; means for calculating a difference between the first and the second sets of errors; and means for encoding the difference.