Object-based digital image predictive coding transfer method and apparatus, and decoding apparatus

Claim 2

An image coding method for coding a target image in a sequence of images, each image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value, said method comprising the steps of: preparing a template which is a representative of the images in sequence; generating a first predictive image by predicting said target image based on said template, and producing first data related to said first predictive image; producing a difference between said first predictive image and said target image; generating a second predictive image by predicting said target image based on a first adjacent image which is to be displayed chronologically adjacent on one side of said target image, and producing second data related to said second predictive image; producing a difference between said second predictive image and said target image; and selecting one of said first and second data that has produced the smallest difference with respect to the target image.

Claim 3

An image coding method according to claim 1, wherein said first adjacent image is chronologically before said target image.

Claim 4

An image coding method according to claim 1, wherein said first adjacent image is chronologically after said target image.

Claim 5

An image coding method according to claim 1, further comprising the steps of: generating a third predictive image by predicting said target image based on a second adjacent image which is to be displayed chronologically adjacent on the other side, opposite to said one side, of said target image, and producing third data related to said third predictive image; producing a difference between said third predictive image and said target image; and said selecting step selecting one of said first, second and third data that has produced the smallest difference with respect to the target image.

Claim 6

An image coding method according to claim 1, further comprising the steps of: generating an average predictive image by obtaining a weighted average of said first and second predictive images, and producing an averaged data related to said average predictive image; producing a difference between said average predictive image and said target image; and said selecting step selecting one of said first, second and averaged data that has produced the smallest difference with respect to the target image.

Claim 7

An image coding method according to claim 4, further comprising the steps of: generating an average predictive image by obtaining a weighted average of at least two of said predictive images, and producing an averaged data related to said average predictive image; producing a difference between said average predictive image and said target image; and said selecting step selecting one of said first, second, third and average data that has produced the smallest difference with respect to the target image.

Claim 8

An image coding method according to claim 6, wherein said at least two of said predictive images are said second and third predictive images.

Claim 9

An image coding method according to claim 1, wherein each of said predictive images is one frame image.

Claim 10

An image coding method according to claim 1, wherein each of said predictive images is a portion of one frame image.

Claim 11

An image predictive coding apparatus comprising: an input means to which an input image is input, said input image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value; a predictor for generating a predictive image; a first adder for generating a difference signal from the input image and the predictive image; a coder for compression coding said difference signal, and outputting a compression coded image; a local decoder for decoding said compression coded image to reproduce said difference signal; a second adder for adding the decoded difference signal to the predictive image to regenerate the reconstructed image; a first frame memory to which the reconstructed image is stored as a template; and a second frame memory to which the reconstructed image is stored as an adjacent image which is adjacent to said input image; said predictor generating a predictive image, using the image stored in the first image memory, when the input image is the template, and generating a predictive image, using the image stored in at least one of the first and second frame memories, when the input image is the adjacent image.

Claim 12

An image predictive coding apparatus comprising: an input means to which an input image is input, said input image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value; a predictor for generating a predictive image; a first adder for generating a difference signal from the input image and the predictive image; a coder for compression coding said difference signal, and outputting a compression coded image; a local decoder for decoding said compression coded image to reproduce said difference signal; a second adder for adding the decoded difference signal to the predictive image to regenerate the reconstructed image; a first frame memory to which the reconstructed image is stored as a template; and a second frame memory to which the reconstructed image is stored as an adjacent image which is adjacent to said input image; said predictor generating a predictive image, using the image stored in the first image memory, when the input image is the template, and generating a predictive image, using the image stored in at least one of the first and second frame memories, when the input image is the adjacent image.

Claim 13

An image coding apparatus for coding a target image in a sequence of images, each image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value, said apparatus comprising: means for preparing a template which is a representative of the images in sequence; means for generating a first predictive image by predicting said target image based on said template, and producing first data related to said first predictive image; means for producing a difference between said first predictive image and said target image; means for generating a second predictive image by predicting said target image based on a first adjacent image which is to be displayed chronologically adjacent on one side of said target image, and producing second data related to said second predictive image; means for producing a difference between said second predictive image and said target image; and means for selecting one of said first and second data that has produced the smallest difference with respect to the target image.

Claim 14

An image coding apparatus for coding a target image in a sequence of images, each image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value, said apparatus comprising: means for preparing a template which is a representative of the images in sequence; means for generating a first predictive image by predicting said target image based on said template, and producing first data related to said first predictive image; means for producing a difference between said first predictive image and said target image; means for generating a second predictive image by predicting said target image based on a first adjacent image which is to be displayed chronologically adjacent on one side of said target image, and producing second data related to said second predictive image; means for producing a difference between said second predictive image and said target image; and means for selecting one of said first and second data that has produced the smallest difference with respect to the target image.

Claim 15

An image coding apparatus according to claim 11, wherein said first adjacent image is chronologically before said target image.

Claim 16

An image coding apparatus according to claim 11, wherein said first adjacent image is chronologically after said target image.

Claim 17

An image coding apparatus according to claim 11, further comprising: means for generating a third predictive image by predicting said target image based on a second adjacent image which is to be displayed chronologically adjacent on the other side, opposite to said one side, of said target image, and producing third data related to said third predictive image; means for producing a difference between said third predictive image and said target image; and said selecting means selecting one of said first, second and third data that has produced the smallest difference with respect to the target image.

Claim 18

An image coding apparatus according to claim 11, further comprising: means for generating an average predictive image by obtaining a weighted average of said first and second predictive images, and producing an averaged data related to said average predictive image; means for producing a difference between said average predictive image and said target image; and said selecting means selecting one of said first, second and averaged data that has produced the smallest difference with respect to the target image.

Claim 19

An image coding apparatus according to claim 14, further comprising: means for generating an average predictive image by obtaining a weighted average of at least two of said predictive images, and producing an averaged data related to said average predictive image; means for producing a difference between said average predictive image and said target image; and said selecting means selecting one of said first, second, third and averaged data that has produced the smallest difference with respect to the target image.

Claim 20

An image coding apparatus according to claim 16, wherein said at least two of said predictive images are said second and third predictive images.

Claim 21

An image coding apparatus according to claim 11, wherein each of said predictive image is one frame image.

Claim 22

An image coding apparatus according to claim 11, wherein each of said predictive image is a portion of one frame image.

Claim 23

A digital image transmission method for transmitting template coded data and coded motion picture data coded from a target image in a sequence of images, each image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value, said method comprising the steps of: preparing a plurality of templates representative of a digitized image sequence; processing said templates separately to make compression coded data as template coded data, and then reproducing said templates as reproduced templates; producing coded motion picture data of a sequence of images through a compression process using one of said templates including generating a first predictive image by predicting said target image based on said template, and producing first data related to said first predictive image; producing a difference between said first predictive image and said target image; generating a second predictive image by predicting said target image based on a first adjacent image which is to be displayed chronologically adjacent on one side of said target image, and producing second data related to said second predictive image; producing a difference between said second predictive image and said target image; and selecting one of said first and second data that has produced the smallest difference with respect to the target image; transmitting said coded motion picture data after first transmitting all of the template coded data.

Claim 24

A digital image transmission method for transmitting template coded data and coded motion picture data coded from a target image in a sequence of images, each image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value, said method comprising the steps of: preparing a plurality of templates representative of a digitized image sequence; processing said templates separately to make compression coded data as template coded data, and then reproducing said templates as reproduced templates; producing coded motion picture data of a sequence of images through a compression process using one of said templates including generating a first predictive image by predicting said target image based on said template, and producing first data related to said first predictive image; producing a difference between said first predictive image and said target image; generating a second predictive image by predicting said target image based on a first adjacent image which is to be displayed chronologically adjacent on one side of said target image, and producing second data related to said second predictive image; producing a difference between said second predictive image and said target image; and selecting one of said first and second data that has produced the smallest difference with respect to the target image; transmitting said coded motion picture data after first transmitting all of the template coded data.

Claim 25

A digital image transmission method according to claim 20, wherein one template coded data and coded motion picture data that uses said one template are transmitted in a group.

Claim 26

An image predictive coding method for predictive coding images in a sequence of images, each image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value, said method comprising the steps of: dividing into a plurality of target subareas the entire area of each target image identified for coding by the corresponding image transmittance signal; determining the luminance signal and transmittance signal of a predictive subarea relative to the luminance signal and transmittance signal of the target subarea; calculating a substitute pixel value for each target subarea containing at least one pixel value not to be coded; generating an optimum predictive subarea by substituting the substitute pixel value into the predictive subarea and the target subarea for the pixel values therein not to be coded; and obtaining and coding the difference between the target subarea and the optimum predictive subarea.

Claim 27

An image predictive coding method for predictive coding images in a sequence of images, each image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value, said method comprising the steps of: dividing into a plurality of target subareas the entire area of each target image identified for coding by the corresponding image transmittance signal; determining the luminance signal and transmittance signal of a predictive subarea relative to the luminance signal and transmittance signal of the target subarea; calculating a substitute pixel value for each target subarea containing at least one pixel value not to be coded; generating an optimum predictive subarea by substituting the substitute pixel value into the predictive subarea and the target subarea for the pixel values therein not to be coded; and obtaining and coding the difference between the target subarea and the optimum predictive subarea.

Claim 28

An image predictive coding method according to claim 22 wherein the calculated substitute pixel value is an average of the pixel value to be coded and contained in the predictive subarea.

Claim 29

An image predictive coding apparatus comprising: an input terminal for receiving an input image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value, a first image block generator for dividing the input image into a target block, a predictor for generating a predictive image from the image stored in a frame memory, a second image block generator for dividing the predictive image into prediction blocks, a padding value generator, upon receipt of the prediction blocks, for calculating a substitute pixel value based on the pixel values to be coded in the prediction block, and outputting a calculated substitute pixel value; a first pixel value padding means, upon receipt of said substitute pixel value and said target block, for substituting the substitute pixel value into the target block for the pixel values therein not to be coded; a second pixel value padding means for receiving the substitute pixel value and prediction block, and for substituting the substitute pixel value into the prediction block for the pixel values therein not to be coded, a first adder for generating a difference signal from the target and prediction blocks in which the substitute pixel value has been substituted, a coder for obtaining compression codes of said difference signal and outputting a compression coded image data; a local decoder for decoding and reproducing the difference signal; a second adder for adding the decoded difference signal to the prediction block to which the substitute pixel value has been substituted to regenerate the reconstructed image, and a frame memory for storing the reconstructed image.

Claim 30

An image predictive coding apparatus comprising: an input terminal for receiving an input image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value, a first image block generator for dividing the input image into a target block, a predictor for generating a predictive image from the image stored in a frame memory, a second image block generator for dividing the predictive image into prediction blocks, a padding value generator, upon receipt of the prediction blocks, for calculating a substitute pixel value based on the pixel values to be coded in the prediction block, and outputting a calculated substitute pixel value; a first pixel value padding means, upon receipt of said substitute pixel value and said target block, for substituting the substitute pixel value into the target block for the pixel values therein not to be coded; a second pixel value padding means for receiving the substitute pixel value and prediction block, and for substituting the substitute pixel value into the prediction block for the pixel values therein not to be coded, a first adder for generating a difference signal from the target and prediction blocks in which the substitute pixel value has been substituted, a coder for obtaining compression codes of said difference signal and outputting a compression coded image data; a local decoder for decoding and reproducing the difference signal; a second adder for adding the decoded difference signal to the prediction block to which the substitute pixel value has been substituted to regenerate the reconstructed image, and a frame memory for storing the reconstructed image.

Claim 31

A decoding apparatus for decoding image data coded by dividing an input image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value, into plural target blocks, determining the luminance signal and transmittance signal of a predictive block relative to the luminance signal and transmittance signal of the target block, calculating a substitute pixel value for each target block containing at least one pixel value not to be coded based on the pixel values to be coded in the corresponding predictive block, substituting the calculated substitute pixel value into the predictive block and the target block for the pixel values therein not to be coded, and then obtaining and coding the difference between the target block and the predictive block, said decoding apparatus comprising: an input terminal receiving said image data; a decoder for generating a difference signal; a frame memory bank, a prediction block generator for generating a prediction block using the image input from the frame memory bank, a padding value generator for receiving the prediction block, and for generating a substitute pixel value based on the pixel values to be coded in the prediction block, a pixel value padding means, upon receipt of the substitute pixel value and the prediction block, for substituting the substitute pixel value into the prediction block for the pixel values therein not to be coded, an adder for adding the substituted prediction block containing the substitute pixel value to the difference signal to generate a reconstructed image, outputting the reconstructed image and storing the reconstructed image to the frame memory bank.

Claim 32

A decoding apparatus for decoding image data coded by dividing an input image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value, into plural target blocks, determining the luminance signal and transmittance signal of a predictive block relative to the luminance signal and transmittance signal of the target block, calculating a substitute pixel value for each target block containing at least one pixel value not to be coded based on the pixel values to be coded in the corresponding predictive block, substituting the calculated substitute pixel value into the predictive block and the target block for the pixel values therein not to be coded, and then obtaining and coding the difference between the target block and the predictive block, said decoding apparatus comprising: an input terminal receiving said image data; a decoder for generating a difference signal; a frame memory bank, a prediction block generator for generating a prediction block using the image input from the frame memory bank, a padding value generator for receiving the prediction block, and for generating a substitute pixel value based on the pixel values to be coded in the prediction block, a pixel value padding means, upon receipt of the substitute pixel value and the prediction block, for substituting the substitute pixel value into the prediction block for the pixel values therein not to be coded, an adder for adding the substituted prediction block containing the substitute pixel value to the difference signal to generate a reconstructed image, outputting the reconstructed image and storing the reconstructed image to the frame memory bank.

Claim 33

A digital image transmission method for transmitting a digital image comprising the steps of: separating an image into a plurality of image layers including a background layer and a foreground layer, said image layers comprising a luminance signal expressing the pixel value, and a transmittance signal expressing the transparency of the pixel value; processing compression coding of each of the plural image layers; multiplexing the layers in a particular sequence without synthesizing the image layers; and transmitting the multiplexed layers.

Claim 34

A digital image transmission method for transmitting a digital image comprising the steps of: separating an image into a plurality of image layers including a background layer and a foreground layer, said image layers comprising a luminance signal expressing the pixel value, and a transmittance signal expressing the transparency of the pixel value; processing compression coding of each of the plural image layers; multiplexing the layers in a particular sequence without synthesizing the image layers; and transmitting the multiplexed layers.

Claim 35

A digital image transmission method according to claim 26, wherein said layers are multiplexed in sequence starting from the background layer.

Claim 36

A digital image transmission method according to claim 26, wherein said layers are multiplexed in sequence starting from the foreground layer.

Claim 39

A digital image decoding and synthesizing apparatus according to claim 29, further comprising: a first selector switch between the external line input terminal and the decoder, said first selector switch being controlled to not connect the external line input terminal and the decoder when an image layer that is not to be reproduced is input.

Claim 40

A digital image decoding and synthesizing apparatus according to claim 30, further comprising: a second selector switch disposed between the synthesizer and the frame memory, said second selector switch being controlled to not connect the synthesizer and the frame memory when the image stored to the frame memory is not to be updated.

Claim 41

A digital image decoding and synthesizing apparatus according to claim 29, further comprising: a first switch; a second switch; a third switch; said frame memory including a first frame memory and a second frame memory, said synthesizer having a first input terminal and a second input terminal, said first input terminal of the synthesizer connected to either one of the decoder and the second frame memory by said first switch, and said second input terminal of the synthesizer connected to either one of the first frame memory and the second frame memory by said second switch, and wherein said new reconstructed image is stored to either one of the first and second frame memories by said third switch.

Claim 42

A digital image decoding and synthesizing apparatus according to claim 29, wherein said coded layers are multiplexed in sequence starting from the background layer.

Claim 43

A digital image decoding and synthesizing apparatus according to claim 29, wherein said coded layers are multiplexed in sequence starting from the foreground layer.

Claim 48

A decoding apparatus for decoding a bit stream of image data coded by dividing an input image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value, into plural target blocks, determining the luminance signal and transmittance signal of a predictive block relative to the luminance signal and transmittance signal of the target block, and then obtaining and coding the difference between the target block and the predictive block, said decoding apparatus comprising: an input terminal receiving the bit stream of image data; a decoder for generating a decoded difference signal; an adder for producing a reconstructed image; a first frame memory to which the reconstructed image is stored as a template; a second frame memory to which the reconstructed image is stored as an adjacent image which is adjusted to the input image; and a predictor for generating a predictive image, using the image stored in the first frame memory, when the input image is the template, and generating a predictive image, using the image stored in at least one of the first and second frame memories, when the input image is the adjacent image; wherein said adder is for adding the decoded difference signal to the predictive image to regenerate the reconstructed image.

Claim 49

A decoding apparatus for decoding a bit stream of image data coded by dividing an input image comprising a luminance signal expressing each pixel value, and a transmittance signal expressing the transparency of each pixel value, into plural target blocks, determining the luminance signal and transmittance signal of a predictive block relative to the luminance signal and transmittance signal of the target block, and then obtaining and coding the difference between the target block and the predictive block, said decoding apparatus comprising: an input terminal receiving the bit stream of image data; a decoder for generating a decoded difference signal; an adder for producing a reconstructed image; a first frame memory to which the reconstructed image is stored as a template; a second frame memory to which the reconstructed image is stored as an adjacent image which is adjusted to the input image; and a predictor for generating a predictive image, using the image stored in the first frame memory, when the input image is the template, and generating a predictive image, using the image stored in at least one of the first and second frame memories, when the input image is the adjacent image; wherein said adder is for adding the decoded difference signal to the predictive image to regenerate the reconstructed image.