Method and apparatus for parametric, model-based, geometric frame partitioning for video coding

An apparatus for video encoding, comprising:

• at least one processor being configured to:
• encode a parametric model for a current block of a picture, wherein said block is partitioned to include a first partition and a second partition according to said parametric model, wherein said at least one processor is configured to partition by:
• associating values with respective pixels of said current block according to said parametric model, and
• classifying said pixels of said current block into partitions according to said associated values, wherein a first value according to said parametric model indicates that a pixel of said current block is in said first partition, a second value indicates that a pixel of said current block is in said second partition, and a value in between said first and second values indicates that a pixel of said current block is in a partial surface;
• predict motion vectors for said first and second partitions based on at least one of spatial neighboring blocks and temporal neighboring blocks;
• encode a first motion vector for said first partition and a second motion vector for said second partition based on said predicted motion vectors;
• predict said current block to form a predicted block by:
• predicting pixels in said first partition using a first predictor,
• predicting pixels in said second partition using a second predictor, and
• predicting pixels in said partial surface based on a weighted linear average of said first predictor and said second predictor, wherein said first predictor is weighted by w(x,y) and said second predictor is weighted by (1−w(x,y)) to form said weighted linear average, wherein w(x, y) is said value from said parametric model for pixel at location (x,y);
• and
• encode said current block based on said predicted block.

A method of video decoding, comprising:

• decoding a parametric model for a current block of a picture, wherein said block is partitioned to include a first partition and a second partition according to said parametric model, wherein said partitioning includes:
• associating values with respective pixels of said current block according to said parametric model, and
• classifying said pixels of said current block into partitions according to said associated values, wherein a first value according to said parametric model indicates that a pixel of said current block is in said first partition, a second value indicates that a pixel of said current block is in said second partition, and a value in between said first and second values indicates that a pixel of said current block is in a partial surface;
• predicting motion vectors for said first and second partitions based on at least one of spatial neighboring blocks and temporal neighboring blocks;
• decoding a first motion vector for said first partition and a second motion vector for said second partition based on said predicted motion vectors;
• predicting said current block to form a predicted block by:
• predicting pixels in said first partition using a first predictor,
• predicting pixels in said second partition using a second predictor, and
• predicting pixels in said partial surface based on a weighted linear average of said first predictor and said second predictor, wherein said first predictor is weighted by w(x,y) and said second predictor is weighted by (1−w(x,y)) to form said weighted linear average, wherein w(x, y) is said value from said parametric model for pixel at location (x,y);
• and
• decoding said current block based on said predicted block.

A method of video encoding, comprising:

• encoding a parametric model for a current block of a picture, wherein said block is partitioned to include a first partition and a second partition according to said parametric model, wherein said partitioning includes:
• associating values with respective pixels of said current block according to said parametric model, and
• classifying said pixels of said current block into partitions according to said associated values, wherein a first value according to said parametric model indicates that a pixel of said current block is in said first partition, a second value indicates that a pixel of said current block is in said second partition, and a value in between said first and second values indicates that a pixel of said current block is in a partial surface;
• predicting motion vectors for said first and second partitions based on at least one of spatial neighboring blocks and temporal neighboring blocks;
• encoding a first motion vector for said first partition and a second motion vector for said second partition based on said predicted motion vectors;
• predicting said current block to form a predicted block by:
• predicting pixels in said first partition using a first predictor,
• predicting pixels in said second partition using a second predictor, and
• predicting pixels in said partial surface based on a weighted linear average of said first predictor and said second predictor, wherein said first predictor is weighted by w(x,y) and said second predictor is weighted by (1−w(x,y)) to form said weighted linear average, wherein w(x, y) is said value from said parametric model for pixel at location (x,y);
• and
• encoding said current block based on said predicted block.