Method and apparatus for encoding a contour image in a video signal

Claim 1

A method for encoding a contour image of an object contained in a video frame of a video signal, wherein the contour image includes a plurality of contour pixels, the contour pixels representing pixels located on the contour image, comprising the steps of: (a) dividing the video frame into a multiplicity of blocks, each of the blocks including K×L pixels with K and L being positive integers, respectively; (b) detecting contour blocks, each of the contour blocks including one or more contour pixels therein; and (c) coding the contour image on a contour block-by-contour block basis; wherein said step (c) includes the steps of: (c1) determining a plurality of vertices on each contour segment included in each contour block, said each contour segment representing a set of continuous contour pixels included in said each contour block; wherein said step (c1) contains the steps of: (c11) determining two vertices on each contour segment, wherein two end points of said each contour segment are selected as said two determined vertices if said each contour segment is of an open loop; and (c12) finding additional vertices on each contour segment until a maximum distance between said each contour segment and a straight line segment connecting two adjacent vertices on said each contour segment becomes smaller than a predetermined threshold value; and (c2) coding the contour image based on the vertices on said each contour segment; wherein said step (c2) contains the steps of: (c21) determining a predetermined number of sample points on a straight line segment joining two adjacent vertices on each contour segment; (c22) calculating an error at each of the sample points and the vertices on each straight line segment, the error representing a distance between the straight line segment and its corresponding contour segment; (c23)transforming a set of errors into a set of transform coefficients; (c24) converting the set of transform coefficients into a set of quantized transform coefficients; and (c25) coding the set of quantized transform coefficients.

Claim 2

A method for encoding a contour image of an object contained in a video frame of a video signal, wherein the contour image includes a plurality of contour pixels, the contour pixels representing pixels located on the contour image, comprising the steps of: (a) dividing the video frame into a multiplicity of blocks, each of the blocks including K×L pixels with K and L being positive integers, respectively; (b) detecting contour blocks, each of the contour blocks including one or more contour pixels therein; and (c) coding the contour image on a contour block-by-contour block basis; wherein said step (c) includes the steps of: (c1) determining a plurality of vertices on each contour segment included in each contour block, said each contour segment representing a set of continuous contour pixels included in said each contour block; wherein said step (c1) contains the steps of: (c11) determining two vertices on each contour segment, wherein two end points of said each contour segment are selected as said two determined vertices if said each contour segment is of an open loop; and (c12) finding additional vertices on each contour segment until a maximum distance between said each contour segment and a straight line segment connecting two adjacent vertices on said each contour segment becomes smaller than a predetermined threshold value; and (c2) coding the contour image based on the vertices on said each contour segment; wherein said step (c2) contains the steps of: (c21) determining a predetermined number of sample points on a straight line segment joining two adjacent vertices on each contour segment; (c22) calculating an error at each of the sample points and the vertices on each straight line segment, the error representing a distance between the straight line segment and its corresponding contour segment; (c23)transforming a set of errors into a set of transform coefficients; (c24) converting the set of transform coefficients into a set of quantized transform coefficients; and (c25) coding the set of quantized transform coefficients.

Claim 5

An apparatus for encoding a contour image of an object contained in a video frame of a video signal, wherein the contour image includes a plurality of contour pixels, the contour pixels representing pixels located on the contour image, which comprises: means for dividing the video frame into a multiplicity of blocks, each of the blocks including K×L pixels with K and L being positive integers, respectively; means for detecting contour blocks from said multiplicity of blocks, each of the contour blocks including one or more contour pixels therein; and means for coding the contour image on a contour block-by-contour block basis; wherein said contour image coding means includes: means for determining a plurality of vertices on each contour segment included in each contour block, said each contour segment representing a set of continuous contour pixels included in each contour block; and wherein said vertices determining means includes: means for determining two vertices on each contour segment, wherein two end points of said each contour segment are selected as said determined two vertices if said each contour segment is of an open loop; and means for finding additional vertices on each contour segment until a maximum distance between said each contour segment and each straight line segment connecting two adjacent vertices on said each contour segment becomes smaller than a predetermined threshold value; and means for coding the contour image based on the vertices on said each contour segment; wherein said contour image coding means includes: means for determining a predetermined number of sample points on each straight line segment joining two adjacent vertices on said each contour segment; means for calculating an error at each of the sample points and the vertices on each straight line segment, the error representing a distance between the straight line segment and its corresponding contour segment; means for transforming a set of errors into a set of transform coefficients; means for converting the set of transform coefficients into a set of quantized transform coefficients; and means for coding the set of quantized transform coefficients.

Claim 6

An apparatus for encoding a contour image of an object contained in a video frame of a video signal, wherein the contour image includes a plurality of contour pixels, the contour pixels representing pixels located on the contour image, which comprises: means for dividing the video frame into a multiplicity of blocks, each of the blocks including K×L pixels with K and L being positive integers, respectively; means for detecting contour blocks from said multiplicity of blocks, each of the contour blocks including one or more contour pixels therein; and means for coding the contour image on a contour block-by-contour block basis; wherein said contour image coding means includes: means for determining a plurality of vertices on each contour segment included in each contour block, said each contour segment representing a set of continuous contour pixels included in each contour block; and wherein said vertices determining means includes: means for determining two vertices on each contour segment, wherein two end points of said each contour segment are selected as said determined two vertices if said each contour segment is of an open loop; and means for finding additional vertices on each contour segment until a maximum distance between said each contour segment and each straight line segment connecting two adjacent vertices on said each contour segment becomes smaller than a predetermined threshold value; and means for coding the contour image based on the vertices on said each contour segment; wherein said contour image coding means includes: means for determining a predetermined number of sample points on each straight line segment joining two adjacent vertices on said each contour segment; means for calculating an error at each of the sample points and the vertices on each straight line segment, the error representing a distance between the straight line segment and its corresponding contour segment; means for transforming a set of errors into a set of transform coefficients; means for converting the set of transform coefficients into a set of quantized transform coefficients; and means for coding the set of quantized transform coefficients.