Method and apparatus for encoding a video signal employing transform and classified vector quantization techniques

Claim 1

A method for encoding an input video signal, comprising the steps of: transforming the input video signal into a plurality of subbands of transform coefficients by using a discrete wavelet transform, wherein one of the subbands existing in the lowest frequency band includes one or more DC coefficients and the remainder of the subbands includes a multiplicity of AC coefficients, respectively; converting the transform coefficients included in the plurality of subbands into a number of assembled blocks with a predetermined identical size, each of the assembled blocks having M×M transform coefficients wherein M is a positive integer, one of the M×M transform coefficients being a DC coefficient and the remainder thereof being. AC coefficients; generating first and second class parameters representing the activities of the DC coefficients in the horizontal and vertical directions within all of the assembled blocks; determining and generating one or more subclasses based on the first and the second class parameters, a predetermined threshold value, and a slope factor; generating a selection signal in response to said one or more subclasses; encoding the DC coefficients in the assembled blocks, to thereby generate a first encoded video signal; encoding the AC coefficients in each of the assembled blocks in response to the selection signal, to thereby produce a second encoded video signal; and formatting the first and the second encoded video signals.

Claim 2

A method for encoding an input video signal, comprising the steps of: transforming the input video signal into a plurality of subbands of transform coefficients by using a discrete wavelet transform, wherein one of the subbands existing in the lowest frequency band includes one or more DC coefficients and the remainder of the subbands includes a multiplicity of AC coefficients, respectively; converting the transform coefficients included in the plurality of subbands into a number of assembled blocks with a predetermined identical size, each of the assembled blocks having M×M transform coefficients wherein M is a positive integer, one of the M×M transform coefficients being a DC coefficient and the remainder thereof being. AC coefficients; generating first and second class parameters representing the activities of the DC coefficients in the horizontal and vertical directions within all of the assembled blocks; determining and generating one or more subclasses based on the first and the second class parameters, a predetermined threshold value, and a slope factor; generating a selection signal in response to said one or more subclasses; encoding the DC coefficients in the assembled blocks, to thereby generate a first encoded video signal; encoding the AC coefficients in each of the assembled blocks in response to the selection signal, to thereby produce a second encoded video signal; and formatting the first and the second encoded video signals.

Claim 3

An apparatus for encoding an input video signal, which comprises: means for transforming the input video signal into a plurality of subbands of transform coefficients by using a discrete wavelet transform, wherein one of the subbands existing in the lowest frequency band includes one or more DC coefficients and the remainder of the subbands includes a multiplicity of AC coefficients, respectively; means for converting the transform coefficients included in the plurality of subbands into a number of assembled blocks with a predetermined identical size, each of the assembled blocks having M×M transform coefficients wherein M is a positive integer, one of the M×M transform coefficients being a DC coefficient and the remainder thereof being AC coefficients; means for generating first and second class parameters representing the activities of the DC coefficients in the horizontal and vertical directions within all of the assembled blocks; means for determining and generating one or more subclasses based on the first and the second class parameters, a predetermined threshold value and a slope factor; means for generating a selection signal in response to the subclasses; first encoding means for encoding the DC coefficients in the assembled blocks, to thereby generate a first encoded video signal; second encoding means, responsive to the selection signal, for encoding the AC coefficients in each of the assembled blocks, to thereby produce a second encoded video signal; and means for formatting the first and the second encoded video signals.

Claim 4

An apparatus for encoding an input video signal, which comprises: means for transforming the input video signal into a plurality of subbands of transform coefficients by using a discrete wavelet transform, wherein one of the subbands existing in the lowest frequency band includes one or more DC coefficients and the remainder of the subbands includes a multiplicity of AC coefficients, respectively; means for converting the transform coefficients included in the plurality of subbands into a number of assembled blocks with a predetermined identical size, each of the assembled blocks having M×M transform coefficients wherein M is a positive integer, one of the M×M transform coefficients being a DC coefficient and the remainder thereof being AC coefficients; means for generating first and second class parameters representing the activities of the DC coefficients in the horizontal and vertical directions within all of the assembled blocks; means for determining and generating one or more subclasses based on the first and the second class parameters, a predetermined threshold value and a slope factor; means for generating a selection signal in response to the subclasses; first encoding means for encoding the DC coefficients in the assembled blocks, to thereby generate a first encoded video signal; second encoding means, responsive to the selection signal, for encoding the AC coefficients in each of the assembled blocks, to thereby produce a second encoded video signal; and means for formatting the first and the second encoded video signals.