# US 8,787,693 B2 · Prediction of images by prior determination of a family of reference pixels, coding and decoding using such a prediction > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 8,787,693 B2 - Document ID: US8787693B2 - Status: Granted - Kind: B2 - Application number: 12/937,293 - Filed: 2009-04-07 - Published: 2011-02-03 - Granted: 2014-07-22 - Earliest claimed: 2008-04-15 - Assignees: Orange - Applicants: Joël Jung; Guillaume Laroche - Inventors: Joël Jung; Guillaume Laroche - Web page: https://patentgrep.com/patent/US8787693B2 ## CPC - H04N 19/649: {the transform being applied to non rectangular image segments} - H04N 19/105: Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction - H04N 19/11: among a plurality of spatial predictive coding modes - H04N 19/119: Adaptive subdivision aspects, e.g. subdivision of a picture into rectangular or non-rectangular coding blocks - H04N 19/147: according to rate distortion criteria (rate-distortion as a criterion for motion estimation H04N19/567) - H04N 19/159: Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction - H04N 19/176: the region being a block, e.g. a macroblock - H04N 19/182: the unit being a pixel - H04N 19/19: using optimisation based on Lagrange multipliers - H04N 19/57: Motion estimation characterised by a search window with variable size or shape - H04N 19/593: involving spatial prediction techniques - H04N 19/61: in combination with predictive coding - H04N 19/70: characterised by syntax aspects related to video coding, e.g. related to compression standards - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION ## Abstract A method is provided for predicting partitions of at least one group of pixels in an image to be coded, with respect to a group of reference pixels. Accordingly, the group of reference pixels is obtained by calculating a function characteristic of a predetermined mode of prediction, the reference pixels of the group obtained being variable from one partition to another. ## Claims ### Claim 1 A method for predicting pixels in an image to be coded with an image coder comprising the steps of: selecting sets of reference pixels; determining distances from a current pixel of a group of current pixels to each pixel of the sets of reference pixels, respectively; selecting a group of reference pixels for the current pixel using a predetermined prediction mode, wherein the group of reference pixels is located in the sets and is closest to the current pixel; predicting the current pixel based on the reference pixel which, out of the group of reference pixels, is closest to the current pixel. ### Claim 2 (depends on 1) The method of claim 1, further comprising: determining the number of reference pixels to be selected for the group of reference pixels, wherein the number corresponds to the predetermined prediction mode. ### Claim 3 (depends on 1) The method of claim 1, further comprising: detecting that a predicted partition contains a contour that prolongs a reference contour, wherein the selecting step comprises selecting at least one reference pixel of the group of reference pixels that is situated in a same position with respect to the reference contour as at least one current pixel of the group of current pixels. ### Claim 4 (depends on 1) The method according to claim 1, further comprising: choosing a distance between at least one of the group of current pixels and at least one of the group of reference pixels, wherein the predetermined prediction mode is based on the chosen distance. ### Claim 5 (depends on 4) The method according to claim 4, wherein the choosing step comprises minimizing the distance so that the reference pixels of the group of reference pixels are situated in a neighborhood closest to the at least one of the group of current pixels. ### Claim 6 (depends on 1) The method according to claim 1, further comprising choosing a direction from at least one current pixel of the group of current pixels to at least one of the group of reference pixels, wherein the predetermined prediction mode is based on the chosen direction. ### Claim 7 (depends on 1) The method according to claim 1, wherein the reference pixels of the group of reference pixels are variable from one partition to another, the method further comprising: predicting, according to the predetermined prediction mode, the partitions of a group of current pixels based on at least one of the group of reference pixels; splitting the group of current pixels into a plurality of partitions according to the predicted partitions; and coding the image using the plurality of partitions. ### Claim 8 (depends on 7) The method according to claim 7, further comprising receiving a data stream that includes data representations of the image; and decoding the image based on the plurality of partitions. ### Claim 9 (depends on 1) A non-transitory computer readable medium comprising instructions for implementing the steps recited in claim 1. ### Claim 10 A method for predicting pixels in an image to be decoded with an image decoder comprising the steps of: selecting sets of reference pixels; determining distances from a current pixel of a group of current pixels to each pixel of the sets of reference pixels, respectively; selecting a group of reference pixels for the current pixel using a predetermined prediction mode, wherein the group of reference pixels is located in the sets and is closest to the current pixel; predicting the current pixel based on the reference pixel which, out of the group of reference pixels, is closest to the current pixel. ### Claim 11 (depends on 10) A non-transitory computer readable medium comprising instructions for implementing the steps recited in claim 10. ### Claim 12 (depends on 10) The method of claim 10, further comprising: determining the number of reference pixels to be selected for the group of reference pixels, wherein the number corresponds to the predetermined prediction mode. ### Claim 13 (depends on 10) The method of claim 10, further comprising: detecting that a predicted partition contains a contour that prolongs a reference contour, wherein the selecting step comprises selecting at least one reference pixel of the group of reference pixels that is situated in a same position with respect to the reference contour as at least one current pixel of the group of current pixels. ### Claim 14 (depends on 10) The method according to claim 10, further comprising: choosing a distance between at least one of the group of current pixels and at least one of the group of reference pixels, wherein the predetermined prediction mode is based on the chosen distance. ### Claim 15 (depends on 14) The method according to claim 14, wherein the choosing step comprises minimizing the distance so that the reference pixels of the group of reference pixels are situated in a neighborhood closest to the at least one of the group of current pixels. ### Claim 16 (depends on 10) The method according to claim 10, further comprising choosing a direction from at least one current pixel of the group of current pixels to at least one of the group of reference pixels, wherein the predetermined prediction mode is based on the chosen direction. ### Claim 17 (depends on 10) The method according to claim 10, wherein the reference pixels of the group of reference pixels are variable from one partition to another, the method further comprising: predicting, according to the predetermined prediction mode, the partitions of a group of current pixels based on at least one of the group of reference pixels; splitting the group of current pixels into a plurality of partitions according to the predicted partitions; and decoding the image using the plurality of partitions. ### Claim 18 (depends on 10) The method of claim 10, wherein the plurality of reference pixels of the group consists of: a first pixel located in the neighborhood immediately above the current pixel; a second pixel located in the neighborhood immediately above the current pixel; a third pixel located in the neighborhood immediately to the left of the current pixel; and a fourth pixel located in the neighborhood immediately to the right of the current pixel. ### Claim 19 A method for predicting pixels in an image to be decoded with an image decoder comprising the steps of: selecting sets of reference pixels; determining distances from a current pixel of a group of current pixels to each pixel of the sets of reference pixels, respectively; selecting a group of reference pixels for the current pixel using a predetermined prediction mode, wherein the group of reference pixels is located in the sets and is closest to the current pixel; predicting the current pixel based on at least one pixel of the group of reference pixels, the at least one pixel being located in a predetermined direction of prediction. ### Claim 20 (depends on 19) The method of claim 19, wherein the predetermined direction of prediction is a horizontal direction. ### Claim 21 (depends on 19) The method of claim 19, wherein the predetermined direction of prediction is a vertical direction. ### Claim 22 (depends on 19) The method of claim 19, wherein the predicting step comprises predicting the current pixel based on a plurality of the group of reference pixels including said at least one pixels, wherein the plurality of pixels are located in the same predetermined direction of prediction. ### Claim 23 A method for predicting pixels in an image to be decoded-with an image decoder comprising the steps of: selecting sets of reference pixels; determining distances from a current pixel of a group of current pixels to each pixel of the sets of reference pixels, respectively; selecting a group of reference pixels for the current pixel using a predetermined prediction mode, the group of references pixels including a first reference pixel and a second reference pixel, wherein the group of reference pixels is located in the sets and is closest to the current pixel, wherein the first reference pixel is located in a first predetermined direction of prediction and the second reference pixel is located in a second predetermined direction of prediction, the first and second predetermined directions being different from one another; predicting the current pixel based on the first and second reference pixels. ### Claim 24 (depends on 23) The method of claim 23, wherein the first predetermined direction of prediction is a vertical direction and the second predetermined direction of prediction is a horizontal direction.