Adaptive Geometric Partitioning For Video Encoding

• US 20090196342A1
• Filed: 07/31/2007
• Published: 08/06/2009
• Est. Priority Date: 08/02/2006
• Status: Abandoned Application

• Current Assignee: Thomson Licensing (Vantiva SA)
• Original Assignee: Thomson Licensing Dtv
• Inventors: Yin, Peng, Divorra Escoda, Oscar
• Application Number:
  US12/309,540
• Publication Number:
  US 20090196342A1
• Time in Patent Office: Days
• US Class Current: 375/240.02
• CPC Class Codes: H04N 19/117 Filters, e.g. for pre-processing or post-processing sub-band filter banks H04N19/635
• CPC Class Codes: H04N 19/119 Adaptive subdivision aspects, e.g. subdivision of a picture into rectangular or non-rectangular coding blocks
• CPC Class Codes: H04N 19/126 Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers
• CPC Class Codes: H04N 19/13 Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC]
• CPC Class Codes: H04N 19/146 Data rate or code amount at the encoder output
• CPC Class Codes: H04N 19/156 Availability of hardware or computational resources, e.g. encoding based on power-saving criteria
• CPC Class Codes: H04N 19/159 Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/189 characterised by the adaptation method, adaptation tool or adaptation type used for the adaptive coding
• CPC Class Codes: H04N 19/44 Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder
• CPC Class Codes: H04N 19/50 using predictive coding H04N19/61 takes precedence
• CPC Class Codes: H04N 19/507 using conditional replenishment
• CPC Class Codes: H04N 19/543 using regions
• CPC Class Codes: H04N 19/57 Motion estimation characterised by a search window with variable size or shape
• CPC Class Codes: H04N 19/61 in combination with predictive coding
• CPC Class Codes: H04N 19/70 characterised by syntax aspects related to video coding, e.g. related to compression standards

Abstract

There are provided methods and apparatus for adaptive geometric partitioning for video encoding and decoding. An apparatus includes an encoder for encoding image data corresponding to pictures by adaptively partitioning at least portions of the pictures responsive to at least one parametric model. The at least one parametric model involves at least one of implicit and explicit formulation of at least one curve.

Claims

1. An apparatus, comprising:

an encoder for encoding image data corresponding to pictures by adaptively partitioning at least portions of the pictures responsive to at least one parametric model, wherein the at least one parametric model involves at least one of implicit and explicit formulation of at least one curve.

2. The apparatus of claim 1, wherein at least one of the at least one parametric model and the at least one curve are derived from a geometric signal model.

3. The apparatus of claim 1, wherein at least one of the at least one parametric model and the at least one curve describe at least one of, one or more image contours, and, one or more motion boundaries.

4. The apparatus of claim 1, wherein at least one polynomial is used as at least one of the at least one parametric model and the at least one curve.

5. The apparatus of claim 1, wherein a first order polynomial model is used as at least one of the at least one parametric model and the at least one curve.

6. The apparatus of claim 5, wherein the first order polynomial model includes an angle parameter and a distance parameter.

7. The apparatus of claim 1, wherein the at least one parametric model for a given image portion is adaptively selected from a set of models when more than one parametric model is available, and the selection is explicitly or implicitly coded.

8. The apparatus of claim 1, wherein said encoder performs explicit or implicit coding of a precision of parameters of at least one of the at least one parametric model and the at least one curve using at least one high level syntax element.

9. The apparatus of claim 8, wherein the at least one high level syntax element is placed at least one of a slice header level, a Supplemental Enhancement Information (SEI) level, a picture parameter set level, a sequence parameter set level and a network abstraction layer unit header level.

10. The apparatus of claim 1, wherein a precision of parameters of at least one of the at least one parametric model and the at least one curve is adapted in order to control at least one of compression efficiency and encoder complexity.

11. The apparatus of claim 10, wherein the precision of the parameters of at least one of the at least one parametric model and the at least one curve is adapted depending on a compression quality parameter.

12. The apparatus of claim 1, wherein predictor data, associated with at least one partition of at least one of the pictures, is predicted from at least one of spatial neighboring blocks and temporal neighboring blocks.

13. The apparatus of claim 1, wherein partition model parameters for at least one of the at least one parametric model and the at least one curve are predicted from at least one of spatial neighboring blocks and temporal neighboring blocks.

14. The apparatus of claim 1, wherein said encoder computes prediction values for pixels that, according to at least one of the at least one parametric model and the at least one curve, lay partly in more than one partition, using at least one of an anti-aliasing procedure, a combination of a part of prediction values for corresponding positions of the pixels, a totality of the prediction values for the corresponding positions of the pixels, a neighborhood, predictors of different partitions, from among the more than one partition, where the pixel is deemed to partly lay.

15. The apparatus of claim 1, wherein said encoder is an extended version of an existing hybrid predictive encoder of an existing video coding standard or video coding recommendation.

16. The apparatus of claim 15, wherein said encoder applies parametric model based partitions to at least one of macroblocks and sub-macroblocks of the pictures as coding modes for at least one of the macroblocks and the sub-macroblocks, respectively.

17. The apparatus of claim 16, wherein parametric model-based coding modes are inserted within existing macroblock and sub-macroblock coding modes of an existing video coding standard or video coding recommendation.

18. The apparatus of claim 16, wherein said encoder encodes model parameters of at least one of the at least one parametric model and the at least one curve to generate the parametric model-based partitions along with partitions prediction data.

19. The apparatus of claim 16, wherein said encoder selects model parameters of at least one of the at least one parametric model, the at least one curve, and partition predictions in order to jointly minimize at least one of a distortion measure and a coding cost measure.

20. The apparatus of claim 16, wherein pixels of at least one of the pictures that overlap at least two parametric model-based partitions are a weighted linear average from predictions of the at least two parametric model-based partitions.

21. The apparatus of claim 16, wherein partition predictions are of at least one of the type inter and intra.

22. The apparatus of claim 16, wherein said encoder selectively uses parameter predictions for at least one of the at least one parametric model and the at least one curve for partition model parameters coding.

23. The apparatus of claim 22, wherein a prediction for a current block of a particular one of the pictures is based on curve extrapolation from neighboring blocks into the current block.

24. The apparatus of claim 22, wherein said encoder uses different contexts or coding tables to encode the image data depending on whether or not parameters of at least one of the at least one parametric model and the at least one curve are predicted.

25. The apparatus of claim 16, wherein said encoder is an extended version of an encoder for the International Organization for Standardization/International Electrotechnical Commission (ISO/IEC) Moving Picture Experts Group-4 (MPEG-4) Part 10 Advanced Video Coding (AVC) standard/International Telecommunication Union, Telecommunication Sector (ITU-T) H.264 recommendation.

26. The apparatus of claim 16, wherein said encoder applies at least one of deblocking filtering and reference frame filtering adapted to handle transform-size blocks affected by at least one parametric model-based partition due to non-tree-based partitioning of the at least one of the macroblocks and the sub-macroblocks when parametric model-based partition modes are used, and wherein the deblocking filtering and the reference frame filtering is dependent upon at least one of whichever one of the at least one parametric model-based partition is used and a selected shape of the at least one parametric model-based partition.

27. The apparatus of claim 15, wherein said encoder adapts at least one of a residual transform and inverse residual transform pair and a quantization procedure de-quantization procedure pair depending on a selected parametric model-based partition.

28. A method, comprising:

encoding image data corresponding to pictures by adaptively partitioning at least portions of the pictures responsive to at least one parametric model, wherein the at least one parametric model involves at least one of implicit and explicit formulation of at least one curve.

29. The method of claim 28, wherein at least one of the at least one parametric model and the at least one curve are derived from a geometric signal model.

30. The method of claim 28, wherein at least one of the at least one parametric model and the at least one curve describe at least one of, one or more image contours, and, one or more motion boundaries.

31. The method of claim 28, wherein at least one polynomial is used as at least one of the at least one parametric model and the at least one curve.

32. The method of claim 28, wherein a first order polynomial model is used as at least one of the at least one parametric model and the at least one curve.

33. The method of claim 32, wherein the first order polynomial model includes an angle parameter and a distance parameter.

34. The method of claim 28, wherein the at least one parametric model for a given image portion is adaptively selected from a set of models when more than one parametric model is available, and the selection is explicitly or implicitly coded.

35. The method of claim 28, wherein said encoding step performs explicit or implicit coding of a precision of parameters of at least one of the at least one parametric model and the at least one curve using at least one high level syntax element.

36. The method of claim 35, wherein the at least one high level syntax element is placed at least one of a slice header level, a Supplemental Enhancement Information (SEI) level, a picture parameter set level, a sequence parameter set level and a network abstraction layer unit header level.

37. The method of claim 28, wherein a precision of parameters of at least one of the at least one parametric model and the at least one curve is adapted in order to control at least one of compression efficiency and encoder complexity.

38. The method of claim 37, wherein the precision of the parameters of at least one of the at least one parametric model and the at least one curve is adapted depending on a compression quality parameter.

39. The method of claim 28, wherein predictor data, associated with at least one partition of at least one of the pictures, is predicted from at least one of spatial neighboring blocks and temporal neighboring blocks.

40. The method of claim 28, wherein partition model parameters for at least one of the at least one parametric model and the at least one curve are predicted from at least one of spatial neighboring blocks and temporal neighboring blocks.

41. The method of claim 28, wherein said encoding step computes prediction values for pixels that, according to at least one of the at least one parametric model and the at least one curve, lay partly in more than one partition, using at least one of an anti-aliasing procedure, a combination of a part of prediction values for corresponding positions of the pixels, a totality of the prediction values for the corresponding positions of the pixels, a neighborhood, predictors of different partitions, from among the more than one partition, where the pixel is deemed to partly lay.

42. The method of claim 28, wherein the encoding step is performed in an encoder that is an extended version of an existing hybrid predictive encoder of an existing video coding standard or video coding recommendation.

43. The method of claim 42, wherein said encoding step applies parametric model based partitions to at least one of macroblocks and sub-macroblocks of the pictures as coding modes for at least one of the macroblocks and the sub-macroblocks, respectively.

44. The method of claim 43, wherein parametric model-based coding modes are inserted within existing macroblock and sub-macroblock coding modes of an existing video coding standard or video coding recommendation.

45. The method of claim 43, wherein said encoding step encodes model parameters of at least one of the at least one parametric model and the at least one curve to generate the parametric model-based partitions along with partitions prediction data.

46. The method of claim 43, wherein said encoding step selects model parameters of at least one of the at least one parametric model, the at least one curve, and partition predictions in order to jointly minimize at least one of a distortion measure and a coding cost measure.

47. The method of claim 43, wherein pixels of at least one of the pictures that overlap at least two parametric model-based partitions are a weighted linear average from predictions of the at least two parametric model-based partitions.

48. The method of claim 43, wherein partitions predictions are of at least one of the type inter and intra.

49. The method of claim 43, wherein said encoding step selectively uses parameter predictions for at least one of the at least one parametric model and the at least one curve for partition model parameters coding.

50. The method of claim 49, wherein a prediction for a current block of a particular one of the pictures is based on curve extrapolation from neighboring blocks into the current block.

51. The method of claim 49, wherein said encoding step uses different contexts or coding tables to encode the image data depending on whether or not parameters of at least one of the at least one parametric model and the at least one curve are predicted.

52. The method of claim 43, wherein said encoding step is performed in an extended version of an encoder for the International Organization for Standardization/International Electrotechnical Commission (ISO/IEC) Moving Picture Experts Group-4 (MPEG-4) Part 10 Advanced Video Coding (AVC) standard/International Telecommunication Union, Telecommunication Sector (ITU-T) H.264 recommendation.

53. The method of claim 43, wherein said encoding step applies at least one of deblocking filtering and reference frame filtering adapted to handle transform-size blocks affected by at least one parametric model-based partition due to non-tree-based partitioning of the at least one of the macroblocks and the sub-macroblocks when parametric model-based partition modes are used, and wherein the deblocking filtering and the reference frame filtering is dependent upon at least one of whichever one of the at least one parametric model-based partition is used and a selected shape of the at least one parametric model-based partition.

54. The method of claim 42, wherein said encoding adapts at least on of a residual transform and inverse residual transform pair and a quantization procedure de-quantization procedure pair depending on a selected parametric model-based partition.

55. A video signal structure for video encoding, comprising:

image data corresponding to pictures encoded by adaptively partitioning at least portions of the pictures responsive to at least one parametric model, wherein the at least one parametric model involves at least one of implicit and explicit formulation of at least one curve.