Method and apparatus for encoding an image signal by using the contour signal thereof

Claim 3

The method according to claim 1, wherein the image signal is either a digital image frame signal or a video object plane signal in a digital image frame signal.

Claim 4

The method according to claim 2, wherein the first binary value is 1 and the second binary value is zero.

Claim 5

The method according to claim 2, wherein the primary padding is carried out by using either a mean padding method or a repetitive padding method.

Claim 6

The method according to claim 4, wherein said step (d) includes the steps of: (d1) dividing the primary padded image signal into the multiplicity of equal-sized primary padded blocks of M×N pixels; (d2) superposing the decoded contour on the primary padded blocks and then redefining pixels located inside the decoded contour as redefined object pixels and pixels located outside the decoded contour as redefined background pixels; and (d3) block defining each of the primary padded blocks as either the contour block or the object block to thereby select the contour blocks and the object blocks.

Claim 7

The method according to claim 5, wherein said step (f) includes the steps of: (f1) designating said each of the contour blocks as either a first contour block or a second contour block to thereby provide first contour blocks as the padded contour blocks and a plurality of second contour blocks based on the mask data, wherein the first contour block is a contour block having no object pixel therein, hence having no contour segment of the object and the second contour block is a contour block having one or more object pixels therein, hence having a contour segment of the object; (f2) searching a corresponding candidate subblock within. each of the masked blocks to thereby provide information representing positions and dimensions of corresponding candidate subblocks, wherein the candidate subblock is a smallest rectangular subblock among subblocks containing all of the first binary values inside the decoded contour within said each of the masked blocks; (f3) forming a minimum subblock within each of the second contour blocks by using the information representing the position and dimension of a corresponding candidate subblock based on the mask data and said each of the second contour blocks to thereby provide a multiplicity of minimum subblocks, wherein each minimum subblock is located at the same position as the corresponding candidate subblock within said each of the second contour blocks, has the same dimension as the corresponding candidate subblock, has a contour segment of the object therein, has the same pixel values as those inside the contour segment of said each of the second contour blocks and has the second binary values as pixel values outside the contour segment therein; (f4) substituting the second binary values outside the contour segment within each of the minimum subblocks with first padded values to form a padded minimum subblock, thereby producing a plurality of padded minimum subblocks; (f5) replacing the second binary values outside each of the padded minimum subblocks with second padded values to form a corresponding padded remaining subblock remained in said each of the second contour blocks, thereby providing a multiplicity of corresponding padded remaining subblocks; and (f6) combining said each of the padded minimum subblocks and the corresponding padded remaining subblock to form a padded contour block, thereby providing a multiplicity of padded contour blocks.

Claim 8

The method according to claim 6, wherein the first padded values are generated by using a distance weighted mean padding method based on the pixel values inside the contour segment in said each of the minimum subblocks.

Claim 9

The method according to claim 7, wherein the second padded values are generated by using a repetitive padding method based on the pixel values in said each of the padded minimum subblocks.

Claim 10

The method according to claim 8, wherein each of the padded minimum subblocks is directly produced as a padded contour block together with the corresponding information representing the position and the dimension of said each of the padded minimum subblocks without going through said steps (f5) and (f6).

Claim 11

The method according to claim 1, wherein said step (g) includes the steps of: (g1) transform coding each of the padded contour blocks produced from said step (f) and each of the object blocks provided from said step (d), thereby providing sets of transform coefficients; (g2) quantizing the sets of transform coefficients, thereby providing sets of quantized transform coefficients; and (g3) entropy coding the sets of quantized transform coefficients, thereby providing the encoded texture signal of the image signal.

Claim 12

An encoding apparatus to encode an image signal based on mask data for the image signal, wherein the image signal includes object pixels within an object and background pixels within a background located outside the object and a contour signal representing the contour of the object and the mask data includes a first binary value used to designate an object pixel and a second binary value used to indicate a background pixel, comprises: means for encoding the contour signal of the object based on the mask data for the image signal, thereby providing an encoded contour signal; means for decoding the encoded contour signal, thereby generating a decoded contour of the object; means for applying primary padding on the image signal, thereby providing a primary padded image signal; means for dividing the primary padded image signal into a multiplicity of equal-sized primary padded blocks of M×N pixels with M and N being predetermined positive integers, respectively, and defining each of the primary padded blocks either as a contour block or as an object block based on the decoded contour to thereby select contour blocks and object blocks, wherein the contour block is a primary padded block having one or more pixels located both and outside the decoded contour and the object block is a primary padded block having pixels located only inside the decoded contour; means for masking each of the contour blocks based on the decoded contour to form each of corresponding masked blocks by replacing every pixel value inside the decoded contour within said each of the contour blocks with the first binary value and every pixel value outside the decoded contour with the second binary value to thereby supply a multiplicity of corresponding masked blocks; means for padding each of the contour blocks based on the mask data and said each corresponding masked blocks to thereby provide a multiplicity of padded contour blocks; and means for encoding the padded contour blocks and the object blocks to thereby provide an encoded texture signal of the image signal.

Claim 13

An encoding apparatus to encode an image signal based on mask data for the image signal, wherein the image signal includes object pixels within an object and background pixels within a background located outside the object and a contour signal representing the contour of the object and the mask data includes a first binary value used to designate an object pixel and a second binary value used to indicate a background pixel, comprises: means for encoding the contour signal of the object based on the mask data for the image signal, thereby providing an encoded contour signal; means for decoding the encoded contour signal, thereby generating a decoded contour of the object; means for applying primary padding on the image signal, thereby providing a primary padded image signal; means for dividing the primary padded image signal into a multiplicity of equal-sized primary padded blocks of M×N pixels with M and N being predetermined positive integers, respectively, and defining each of the primary padded blocks either as a contour block or as an object block based on the decoded contour to thereby select contour blocks and object blocks, wherein the contour block is a primary padded block having one or more pixels located both and outside the decoded contour and the object block is a primary padded block having pixels located only inside the decoded contour; means for masking each of the contour blocks based on the decoded contour to form each of corresponding masked blocks by replacing every pixel value inside the decoded contour within said each of the contour blocks with the first binary value and every pixel value outside the decoded contour with the second binary value to thereby supply a multiplicity of corresponding masked blocks; means for padding each of the contour blocks based on the mask data and said each corresponding masked blocks to thereby provide a multiplicity of padded contour blocks; and means for encoding the padded contour blocks and the object blocks to thereby provide an encoded texture signal of the image signal.

Claim 14

The apparatus according to claim 11, wherein the image signal is either a digital image frame signal or a video object plane signal in a digital image frame signal.

Claim 15

The apparatus according to claim 12, wherein the first binary value is 1 and the second binary value is zero.

Claim 16

The apparatus according to claim 12, wherein the primary padding is carried out by using either a mean padding method or a repetitive padding method.

Claim 17

The apparatus according to claim 14, wherein said dividing means includes: means for dividing the primary padded image signal into the multiplicity of equal-sized primary padded blocks of M×N pixels; means for superposing the decoded contour on the primary padded blocks and then redefining pixels located inside the decoded contour as redefined object pixels and pixels located outside the decoded contour as redefined background pixels; and means for block defining each of the primary padded blocks as either the contour block or the object block to thereby select the contour blocks and the object blocks.

Claim 18

The apparatus according to claim 15, wherein said padding means includes: means for designating said each of the contour blocks as either a first contour block or a second contour block to thereby provide first contour blocks as the padded contour blocks and a plurality of second contour blocks based on the mask data, wherein the first contour block is a contour block having no object pixel therein, hence having no contour segment of the object and the second contour block is a contour block having one or more object pixels therein, hence having a contour segment of the object; means for searching a corresponding candidate subblock within each of the masked blocks to thereby provide information representing positions and dimensions of corresponding candidate subblocks, wherein the candidate subblock is a smallest rectangular subblock among subblocks containing all of the first binary values inside the decoded contour within said each of the masked blocks; means for forming a minimum subblock within each of the second contour blocks by using the information representing the position and dimension of a corresponding candidate subblock based on the mask data and said each of the second contour blocks to thereby provide a multiplicity of minimum subblocks, wherein each minimum subblock is located at the same position as the corresponding candidate subblock within said each of the second contour blocks, has the same dimension as the corresponding candidate subblock, has a contour segment of the object therein, has the same pixel values as those inside the contour segment of said each of the second contour blocks and has the second binary values as pixel values outside the contour segment therein; means for substituting the second binary values outside the contour segment within each of the minimum subblocks with first padded values to form a padded minimum subblock, thereby producing a plurality of padded minimum subblocks; means for replacing the second binary values outside each of the padded minimum subblocks with second padded values to form a corresponding padded remaining subblock remained in said each of the second contour blocks, thereby providing a multiplicity of corresponding padded remaining subblocks; and means for combining said each of the padded minimum subblocks and the corresponding padded remaining subblock to form a padded contour block, thereby providing a multiplicity of padded contour blocks.

Claim 19

The apparatus according to claim 16, wherein the first padded values are generated by using a distance weighted mean padding method based on the pixel values inside the contour segment in said each of the minimum subblocks.

Claim 20

The apparatus according to claim 17, wherein the second padded values are generated by using a repetitive padding method based on the pixel values in said each of the padded minimum subblocks.

Claim 21

The apparatus according to claim 11, wherein said means for encoding the padded contour blocks and the object blocks includes: means for transform coding each of the padded contour blocks and each of the object blocks, thereby providing sets of transform coefficients; means for quantizing the sets of transform coefficients, thereby providing sets of quantized transform coefficients; and means for entropy coding the sets of quantized transform coefficients, thereby providing the encoded texture signal of the image signal.