Integer transform video compression system, method and computer program product
AssigneeCisco Technology, Inc.
InventorsGisle Bjøntegaard, Arild Fuldseth
Life of this patent
Estimating the term…
Abstract
A video coding/decoding system, method and computer program product employ an integer transform matrix for transforming to/from transform coefficients and residual pixel data in moving pictures by a set of semi-orthonormal basis vectors. The basis vectors are derived from conventional DCT or KTL matrixes, but relaxes to some extent the requirements for orthogonality, norm equality and element size limitation. In this way improved coding efficiency and lower complexity compared to previously used integer transforms are possible.
Claims28
A video decoding method comprising:
- inverse transforming with a processing device a first block of transform coefficients by a transform matrix having a plurality of basis vectors into a corresponding second block of residual pixel values for a video image,
- wherein the transform matrix
- consists of integer elements,
- has basis vectors derived from one of a DCT and a KLT,
- has a DC vector,
- not all of the basis vectors being orthogonal but all of the basis vectors being within a predetermined deviation of the DC vector,
- norms of the basis vectors having respective deviations with respect to said DC vector less than 0.0015, and at least one norm-deviation being non-zero,
- having a maximum element deviation from said one of a DCT and a KLT less than or equal to 0.05, and at least one element deviation being non-zero, and each integer element being represented in one byte.
A video decoder comprising:
- a non-transitory computer readable storage device having stored therein a transform matrix having a plurality of basis vectors;
- and
- a processing device configured to inverse transform with a first block of transform coefficients by the transform matrix into a corresponding second block of residual pixel values for a video image,
- wherein the transform matrix
- consists of integer elements,
- has basis vectors derived from one of a DCT and a KLT,
- has a DC vector,
- not all of the basis vectors being orthogonal but all of the basis vectors being within a predetermined deviation of the DC vector,
- norms of the basis vectors having respective deviations with respect to said DC vector less than 0.0015, and at least one norm-deviation being non-zero,
- having a maximum element deviation from said one of a DCT and a KLT less than or equal to 0.05, and at least one element deviation being non-zero, and each integer element being represented in one byte.
A non-transitory computer readable storage device having computer readable instructions stored therein that when executed by a processor perform a video decoder method comprising:
- inverse transforming with a processing device a first block of transform coefficients by a transform matrix having a plurality of basis vectors into a corresponding second block of residual pixel values for a video image,
- wherein the transform matrix consists of integer elements,
- has basis vectors derived from one of a DCT and a KLT,
- has a DC vector,
- not all of the basis vectors being orthogonal but all of the basis vectors being within a predetermined deviation of the DC vector,
- norms of the basis vectors having respective deviations with respect to said DC vector less than 0.0015, and at least one norm-deviation being non-zero,
- having a maximum element deviation from said one of a DCT and a KLT less than or equal to 0.05, and at least one element deviation being non-zero, and each integer element being represented in one byte.
A method of video coding comprising:
- transforming with a processing device a first block of residual pixel values of an image by a transform matrix having a plurality of basis vectors into a corresponding second block of transform coefficients,
- wherein the transform matrix
- consists of integer elements,
- has basis vectors derived from one of a DCT and a KLT,
- has a DC vector,
- not all of the basis vectors being orthogonal but all of the basis vectors being within a predetermined deviation of the DC vector,
- norms of the basis vectors having respective deviations with respect to said DC vector less than 0.0015, and at least one norm-deviation being non-zero,
- having a maximum element deviation from said one of a DCT and a KLT less than or equal to 0.05, and at least one element deviation being non-zero, and each integer element being represented in one byte.
Dependencies are inferred from the claim text.