Vector quantization

Claim 2

An automatic method in an encoding device of vector quantization of an image comprising the following steps: a. initializing N initial nodes in a vector quantizer tree in a dynamic storage device of said encoding device; b. sampling a vector from said image; c. determining a node in said vector quantizer tree which is a best representative sample of the vector sampled from said image; d. associating the vector with said node in said vector quantizer tree in said dynamic storage device; e. sampling a next vector from said image; f. repeating steps c-f until there are no more vectors to be sampled from said image, said next vector becoming said vector; g. determining which of the nodes in said tree is the most distorted node in said tree; h. splitting said most distorted node into two children nodes in said dynamic storage device; i. associating a first portion of the vectors associated with said most distorted node with a first of said children nodes in said dynamic storage device, and a second portion of the vectors associated with said most distorted node with a second of said children nodes in said dynamic storage device; j. determining a current error of the two children nodes compared to the first and second portions of the vectors; k. if the change in error between the current error and a previous error is less than an error threshold then proceeding to step l otherwise determining new values of said first and second children, and proceeding to step i, said current error becoming said previous error; l. repeating steps g through l until the number of terminal nodes in said vector quantizer tree has reached a desired population; and m. associating indices with each of the terminal nodes in said vector quantizer tree in said dynamic storage device.

Claim 3

The method of claim 1 further comprising the step of transmitting a sequence of indices from said vector quantizer tree representative of an index of a terminal node in said vector quantizer tree associated with each said sampled vector in said input image.

Claim 4

The method of claim 1 wherein the step of determining the most distorted node comprises determining whether the average distortion of the node compared to each of vectors sampled from said input image associated with the node has exceeded a threshold.

Claim 5

The method of claim 1 wherein the step of determining the most distorted node comprises determining whether the total distortion of the node compared to each of the vectors sampled from said input image associated with the node has exceeded a threshold.

Claim 6

The method of claim 1 wherein the step of determining the most distorted node comprises determining whether the total population of sampled vectors from said input image associated with said node has exceeded a threshold.

Claim 7

The method of claim 1 wherein the step of determining the most distorted node comprises determining whether the percentage distortion of the node compared to each of the vectors sampled from said input image associated with the node has exceeded a threshold.

Claim 8

The method of claim 1 wherein the step of determining the most distorted node comprises determining whether the maximum distortion of the node compared to each of the vectors sampled from said input image associated with the node has exceeded a threshold.

Claim 9

The method of claim 1 wherein the step of determining the most distorted node comprises determining whether the ratio of maximum to minimum distortion of the node compared to each of the vectors sampled from said input image associated with the node has exceeded a threshold.

Claim 10

The method of claim 1 wherein the step of determining the node which is the best representative sample of the vector sampled from said image comprises determining the mean squared error between the sampled vector and the node.

Claim 11

The method of claim 9 wherein the step of determining the mean squared error is weighed more heavily towards large errors during an early portion of said vector quantization, and weighed less heavily towards large errors during a latter portion of said vector quantization.

Claim 12

The method of claim 1 wherein the step of creating N initial nodes comprises using N initial nodes from a previous vector quantization which has been performed on a previous image.

Claim 13

The method of claim 1 wherein further comprising the additional step of adding a pseudo-randomly generated value to the vector sample from said input image prior to splitting said most distorted node.

Claim 14

The method of claim 1 wherein the step of determining the node which is the best representative sample of the vector sampled from said image comprises determining the node which has luminance and chrominance (YUV) values closest to the sampled vector from the image.

Claim 15

The method of claim 1 further comprising the step of generating a separate vector quantizer tree for different zones in said image.

Claim 16

The method of claim 14 further comprising a step of determining different zones in said image which have variable sizes.

Claim 17

The method of claim 1 which is applied to a sequence of images, the creation of a new vector quantizer tree being performed when a scene change is detected in said sequence of images.

Claim 18

An automatic apparatus for vector quantization of an image comprising: a. means for initializing N initial nodes in a vector quantizer tree in a dynamic storage device of said encoding device; b. means for sampling a vector from said image; c. means for determining a node in said vector quantizer tree which is a best representative sample of the vector sampled from said image; d. means for associating the vector with said node in said vector quantizer tree in said dynamic storage device; e. means for sampling a next vector from said image; f. means for activating components c-e until there are no more vectors remain to be sampled from said image, said next vector becoming said vector; g. means for determining which of the nodes in said tree is the most distorted node in said tree; h. means for splitting said most distorted node into two children nodes in said dynamic storage device; i. means for associating a first portion of the vectors associated with said most distorted node with a first of said children nodes in said dynamic storage device, and a second portion of the vectors associated with said most distorted node with a second of said children nodes in said dynamic storage device; j. means for determining a current error of the two children nodes relative to the first and second portions of the vectors; k. means for determining new values of said first and second children nodes and continuously activating components i-j if the change in error between the current error and a previous error is greater than an error threshold said current error becoming said previous error; l. means for continuously activating components g through l until the number of terminal nodes in said vector quantizer tree has reached a desired population; and m. means for associating indices with each of the terminal nodes in said vector quantizer tree in said dynamic storage device.

Claim 19

An automatic apparatus for vector quantization of an image comprising: a. means for initializing N initial nodes in a vector quantizer tree in a dynamic storage device of said encoding device; b. means for sampling a vector from said image; c. means for determining a node in said vector quantizer tree which is a best representative sample of the vector sampled from said image; d. means for associating the vector with said node in said vector quantizer tree in said dynamic storage device; e. means for sampling a next vector from said image; f. means for activating components c-e until there are no more vectors remain to be sampled from said image, said next vector becoming said vector; g. means for determining which of the nodes in said tree is the most distorted node in said tree; h. means for splitting said most distorted node into two children nodes in said dynamic storage device; i. means for associating a first portion of the vectors associated with said most distorted node with a first of said children nodes in said dynamic storage device, and a second portion of the vectors associated with said most distorted node with a second of said children nodes in said dynamic storage device; j. means for determining a current error of the two children nodes relative to the first and second portions of the vectors; k. means for determining new values of said first and second children nodes and continuously activating components i-j if the change in error between the current error and a previous error is greater than an error threshold said current error becoming said previous error; l. means for continuously activating components g through l until the number of terminal nodes in said vector quantizer tree has reached a desired population; and m. means for associating indices with each of the terminal nodes in said vector quantizer tree in said dynamic storage device.

Claim 20

An automatic apparatus in an encoding device for vector quantization of an image comprising: a. means for initializing N initial nodes in a vector quantizer tree in a dynamic storage device of said encoding device wherein N is greater than 2; b. means for sampling non-overlapping fixed-size vectors from said image; c. means for determining nodes in said vector quantizer tree which are best representative samples of the vectors sampled from said image; d. means for associating said vectors with said nodes in said vector quantizer tree in said dynamic storage device; e. means for iterating and creating new nodes in the vector quantizer tree in said dynamic storage device by determining worst nodes in said tree, splitting said nodes into more than two children nodes and reassociating said vectors with said children nodes in said vector quantizer tree until a number of terminal nodes in said tree reaches a desired population.

Claim 21

An automatic apparatus in an encoding device for vector quantization of an image comprising: a. means for initializing N initial nodes in a vector quantizer tree in a dynamic storage device of said encoding device wherein N is greater than 2; b. means for sampling non-overlapping fixed-size vectors from said image; c. means for determining nodes in said vector quantizer tree which are best representative samples of the vectors sampled from said image; d. means for associating said vectors with said nodes in said vector quantizer tree in said dynamic storage device; e. means for iterating and creating new nodes in the vector quantizer tree in said dynamic storage device by determining worst nodes in said tree, splitting said nodes into more than two children nodes and reassociating said vectors with said children nodes in said vector quantizer tree until a number of terminal nodes in said tree reaches a desired population.

Claim 22

A tree-searched vector quantizer for encoding an image, said tree-searched vector quantizer for generating a codebook from first signals received from said image, said first signals representative of vectors sampled from said image, and terminal nodes in a tree created by said tree-searched vector quantizer are used for said codebook, wherein second signals are transmitted from said vector quantizer, said second signals including indices referencing said terminal nodes in said tree, said tree-searched vector quantizer comprising: a processor; and a storage device coupled to said processor; said storage device having stored therein executable code which, when executed by said processor, causes said processor to perform the steps of: a. determining which of the nodes in said tree is a most distorted node in said tree based on error criteria; b. splitting said most distorted node into children nodes by performing the following steps: sampling K representative vectors for said children nodes; perturbing vectors for said most distorted node in order to create said children nodes if said step of sampling K representative vectors fails to split said most distorted node; re-associating vectors from said most distorted node with said children nodes; and tagging said most distorted node as a terminal node if said children nodes are unable to be created; c. repeating steps a and b until a desired population of said terminal nodes in said tree are created.

Claim 23

A tree-searched vector quantizer for encoding an image, said tree-searched vector quantizer for generating a codebook from first signals received from said image, said first signals representative of vectors sampled from said image, and terminal nodes in a tree created by said tree-searched vector quantizer are used for said codebook, wherein second signals are transmitted from said vector quantizer, said second signals including indices referencing said terminal nodes in said tree, said tree-searched vector quantizer comprising: a processor; and a storage device coupled to said processor; said storage device having stored therein executable code which, when executed by said processor, causes said processor to perform the steps of: a. determining which of the nodes in said tree is a most distorted node in said tree based on error criteria; b. splitting said most distorted node into children nodes by performing the following steps: sampling K representative vectors for said children nodes; perturbing vectors for said most distorted node in order to create said children nodes if said step of sampling K representative vectors fails to split said most distorted node; re-associating vectors from said most distorted node with said children nodes; and tagging said most distorted node as a terminal node if said children nodes are unable to be created; c. repeating steps a and b until a desired population of said terminal nodes in said tree are created.

Claim 24

The tree-searched vector quantizer of claim 19 wherein said step of determining which of the nodes in said tree is the most distorted based on error criteria comprises the step of determining which of the nodes in said tree is the most distorted based on a total distortion of each of the nodes.

Claim 25

The tree-searched vector quantizer of claim 19 wherein said step of determining which of the nodes in said tree is the most distorted based on error criteria comprises the step of determining which of the nodes in said tree is the most distorted based on average distortion of each of the nodes.

Claim 26

The tree-searched vector quantizer of claim 19 wherein said step of determining which of the nodes in said tree is the most distorted based on error criteria comprises the step of determining which of the nodes in said tree is the most distorted based on vector populations associated with each of the nodes.

Claim 27

The tree-searched vector quantizer of claim 19 wherein said step of determining which of the nodes in said tree is the most distorted based on error criteria comprises the step of determining which of the nodes in said tree is the most distorted based on distortion percentages associated with each of the nodes.

Claim 28

The tree-searched vector quantizer of claim 19 wherein said step of determining which of the nodes in said tree is the most distorted based on error criteria comprises the step of determining which of the nodes in said tree is the most distorted based on a maximum distortion of the nodes.

Claim 29

The tree-searched vector quantizer of claim 19 wherein said step of determining which of the nodes in said tree is the most distorted based on error criteria comprises the step of determining which of the nodes in said tree is the most distorted based on a ratio of distortions of the nodes.

Claim 30

An encoding apparatus for compressing an image by performing vector quantization on said image, said apparatus comprising: a processor; and a storage device coupled to said processor; said storage device having stored therein executable code which, when executed by said processor, causes said processor to perform the steps of: a. initializing N initial nodes in a vector quantizer tree; b. sampling a vector from said image; c. determining a node in said vector quantizer tree which is a best representative sample of the vector sampled from said image; d. associating the vector with said node in said vector quantizer tree; e. sampling a next vector from said image; f. repeating steps c-f until there are no more vectors to be sampled from said image, said next vector becoming said vector; g. determining which of the nodes in said tree is the most distorted node in said tree; h. splitting said most distorted node into two children nodes; i. associating a first portion of the vectors associated with said most distorted node with a first of said children nodes, and a second portion of the vectors associated with said most distorted node with a second of said children nodes; j. determining a current error of the two children nodes compared to the first and second portions of the vectors; k. if the change in error between the current error and a previous error is less than an error threshold then proceeding to step 1 otherwise determining new values of said first and second children, and proceeding to step i, said current error becoming said previous error; l. repeating steps g through k until the number of terminal nodes in said vector quantizer tree has reached a desired population; and m. associating indices with each of the terminal nodes in said vector quantizer tree.

Claim 31

An encoding apparatus for compressing an image by performing vector quantization on said image, said apparatus comprising: a processor; and a storage device coupled to said processor; said storage device having stored therein executable code which, when executed by said processor, causes said processor to perform the steps of: a. initializing N initial nodes in a vector quantizer tree; b. sampling a vector from said image; c. determining a node in said vector quantizer tree which is a best representative sample of the vector sampled from said image; d. associating the vector with said node in said vector quantizer tree; e. sampling a next vector from said image; f. repeating steps c-f until there are no more vectors to be sampled from said image, said next vector becoming said vector; g. determining which of the nodes in said tree is the most distorted node in said tree; h. splitting said most distorted node into two children nodes; i. associating a first portion of the vectors associated with said most distorted node with a first of said children nodes, and a second portion of the vectors associated with said most distorted node with a second of said children nodes; j. determining a current error of the two children nodes compared to the first and second portions of the vectors; k. if the change in error between the current error and a previous error is less than an error threshold then proceeding to step 1 otherwise determining new values of said first and second children, and proceeding to step i, said current error becoming said previous error; l. repeating steps g through k until the number of terminal nodes in said vector quantizer tree has reached a desired population; and m. associating indices with each of the terminal nodes in said vector quantizer tree.

Claim 32

The encoding apparatus of claim 26 wherein the step of determining the most distorted node comprises determining whether the average distortion of the node compared to each of vectors sampled from said image associated with the node has exceeded a threshold.

Claim 33

The encoding apparatus of claim 26 wherein the step of determining the most distorted node comprises determining whether the total distortion of the node compared to each of the vectors sampled from said image associated with the node has exceeded a threshold.

Claim 34

The encoding apparatus of claim 26 wherein the step of determining the most distorted node comprises determining whether the total population of sampled vectors from said image associated with said node has exceeded a threshold.

Claim 35

The encoding apparatus of claim 26 wherein the step of determining the most distorted node comprises determining whether the percentage distortion of the node compared to each of the vectors sampled from said image associated with the node has exceeded a threshold.

Claim 36

The encoding apparatus of claim 26 wherein the step of determining the most distorted node comprises determining whether the maximum distortion of the node compared to each of the vectors sampled from said image associated with the node has exceeded a threshold.

Claim 37

The encoding apparatus of claim 26 wherein the step of determining the most distorted node comprises determining whether the ratio of maximum to minimum distortion of the node compared to each of the vectors sampled from said image associated with the node has exceeded a threshold.

Claim 38

The encoding apparatus of claim 26 wherein the step of determining the most distorted node comprises determining whether multiple error criteria compared to each of the vectors sampled from said image associated with the node has exceeded a threshold.

Claim 39

The encoding apparatus of claim 26 wherein the step of determining the node which is the best representative sample of the vector sampled from said image comprises determining the mean squared error between the sampled vector and the node.

Claim 40

The encoding apparatus of claim 34 wherein the step of determining the mean squared error is weighed more heavily towards large errors during an early portion of said vector quantization, and weighed less heavily towards large errors during a latter portion of said vector quantization.

Claim 41

The encoding apparatus of claim 26 wherein the step of creating N initial nodes comprises using N initial nodes from a previous vector quantization which has been performed on a previous image.

Claim 42

The encoding apparatus of claim 28 wherein the vector quantization further comprises the additional step of adding a pseudo-randomly generated value to the vector sample from said image prior to splitting said most distorted node.

Claim 43

The encoding apparatus of claim 26 wherein the step of determining the node which is the best representative sample of the vector sampled from said image comprises determining the node which has luminance and chrominance (YUV) values closest to the sampled vector from the image.

Claim 44

The encoding apparatus of claim 26 wherein the vector quantization further comprises the step of generating a separate vector quantizer tree for different zones in said image.

Claim 45

The encoding apparatus of claim 39 wherein the vector quantization further comprises a step of determining different zones in said image which have variable sizes.

Claim 46

The encoding apparatus of claim 26 wherein the vector quantization further comprises is applied to a sequence of images, the creation of a new vector quantizer tree being performed when a scene change is detected in said sequence of images.