METHOD AND DEVICE FOR ARITHMETIC ENCODING OR ARITHMETIC DECODING

• US 20120195375A1
• Filed: 10/01/2010
• Published: 08/02/2012
• Est. Priority Date: 10/09/2009
• Status: Active Grant

• Current Assignee: Dolby Laboratories Licensing Corporation (Dolby Laboratories Incorporated)
• Original Assignee: Dolby Laboratories Licensing Corporation (Dolby Laboratories Incorporated)
• Inventors: Wuebbolt, Oliver
• Granted Patent: US 9,219,498 B2
• Time in Patent Office: Days
• US Class Current: 375/240.05
• CPC Class Codes: H03M 7/4006 Conversion to or from arithmetic code
• CPC Class Codes: H04N 19/124 Quantisation
• CPC Class Codes: H04N 19/182 the unit being a pixel

Abstract

The invention proposes a method and a device for arithmetic encoding of a current spectral coefficient using preceding spectral coefficients. Said preceding spectral coefficients are already encoded and both, said preceding and current spectral coefficients, are comprised in one or more quantized spectra resulting from quantizing time-frequency—transform of video, audio or speech signal sample values. Said method comprises processing the preceding spectral coefficients, using the processed preceding spectral coefficients for determining a context class being one of at least two different context classes, using the determined context class and a mapping from the at least two different context classes to at least two different probability density functions for determining the probability density function, and arithmetic encoding the current spectral coefficient based on the determined probability density function wherein processing the preceding spectral coefficients comprises non-uniformly quantizing absolutes of the preceding spectral coefficients for use in determining of the context class.

Claims

1-15. -15. (canceled)

16. A method for arithmetic decoding of a current spectral coefficient using preceding spectral coefficients, said preceding spectral coefficients are already decoded and both, said preceding and current spectral coefficients, are comprised in one or more quantized spectra resulting from quantizing time-frequency-transform of video, audio or speech signal sample values, said method comprising using processing means for processing the preceding spectral coefficients, using the processed preceding spectral coefficients for determining a context class being one of at least two different context classes, using the determined context class and a mapping from the at least two different context classes to at least two different probability density functions for determining the probability density function, and arithmetic decoding the current spectral coefficient based on the determined probability density function wherein processing the preceding spectral coefficients comprises non-uniformly quantizing absolutes of the preceding spectral coefficients for use in determination of the context class.

17. The method of claim 16, wherein a sum of the quantized absolutes of the preceding spectral coefficients is used for in determination of the context class.

18. The method of claim 16, wherein processing the preceding spectral coefficients comprises a first quantizing step in which the absolutes of the preceding spectral coefficients are quantized according a first quantization scheme, a step of determining a variance of the absolutes of the preceding spectral coefficients quantized according the first quantization scheme, a step of using the determined variance for selecting one of at least two different nonlinear second quantization schemes, and a second quantizing step in which the absolutes of the preceding spectral coefficients quantized according the first quantization scheme are further quantized according to the selected nonlinear second quantization scheme.

19. The method of claim 18, wherein the step of determining the variance of the absolutes of the preceding spectral coefficients quantized according the first quantization scheme comprises determining a sum of the absolutes of the preceding spectral coefficients quantized according the first quantization scheme and comparing the determined sum with at least one threshold.

20. The method of claim 16, wherein processing either results in a first outcome or at least a different second outcome and determining the context class further comprises determining a number of preceding spectral coefficients being non-uniformly quantized to the first outcome, and using the determined number for determining the context class.

21. The method of claim 20, wherein one of the preceding spectral coefficients is preferred over the preceding spectral coefficients remaining, said method further comprises using the non-uniformly quantization of the preferred one among the preceding spectral coefficient for determining the context class.

22. The method of claim 21, wherein said preferred among the preceding spectral coefficients is comprised in a preceding spectrum and said current spectral coefficient is comprised in a different current spectrum, said preferred preceding and said current spectral coefficients being comprised at a same frequency in the respective spectrum.

23. The method of claim 22, wherein said different current spectrum further comprises at least one of the remaining preceding spectral coefficients at a frequency lower than said same frequency.

24. The method of claim 16, said method further comprises receiving at least one of a mode switching signal and a reset signal and using the at least one received signal for controlling the step of determining the context class.

25. The method of claim 16, said method further comprises using a representative set of data for determining the at least two different probability density functions.

26. The method of claim 16, wherein the mapping is realized using a look-up table or a hash table.

29. A non-transitory storage medium carrying arithmetic encoded spectral coefficients arithmetic encoded according to the method of claim 17.

27. A device for arithmetic encoding of a current spectral coefficient using preceding spectral coefficients, said preceding spectral coefficients are already encoded and both, said preceding and current spectral coefficients, are comprised in one or more quantized spectra resulting from quantizing time-frequency-transform of video, audio or speech signal sample values, said device comprising processing means for processing the preceding spectral coefficients, first means for determining a context class being one of at least two different context classes, said first means being adapted for using the processed preceding spectral coefficients for determining the context class, second means for determining the probability density function, said second means being adapted for using the determined context class and a mapping from the at least two different context classes to at least two different probability density functions for determining the probability density function, and an arithmetic encoder for arithmetic encoding the current spectral coefficient based on the determined probability density function wherein the processing means are adapted for non-uniformly quantizing absolutes of the preceding spectral coefficients for use in determination of the context class.

28. A device for arithmetic decoding of a current spectral coefficient using preceding spectral coefficients, said preceding spectral coefficients are already decoded and both, said preceding and current spectral coefficients, are comprised in one or more quantized spectra resulting from quantizing time-frequency-transform of video, audio or speech signal sample values, said device comprising processing means for processing the preceding spectral coefficients, first means for determining a context class being one of at least two different context classes, said first means being adapted for using the processed preceding spectral coefficients for determining the context class, second means for determining the probability density function, said second means being adapted for using the determined context class and a mapping from the at least two different context classes to at least two different probability density functions for determining the probability density function, and an arithmetic decoder for arithmetic decoding the current spectral coefficient based on the determined probability density function wherein the processing means are adapted for non-uniformly quantizing absolutes of the preceding spectral coefficients for use in determination of the context class.

30. A method for arithmetic encoding of a current spectral coefficient using preceding spectral coefficients, said preceding spectral coefficients are already encoded and both, said preceding and current spectral coefficients, are comprised in one or more quantized spectra resulting from quantizing time-frequency-transform of video, audio or speech signal sample values, said method comprising using processing means for processing the preceding spectral coefficients, using the processed preceding spectral coefficients for determining a context class being one of at least two different context classes, using the determined context class and a mapping from the at least two different context classes to at least two different probability density functions for determining the probability density function, and arithmetic encoding the current spectral coefficient based on the determined probability density function wherein processing the preceding spectral coefficients comprises non-uniformly quantizing absolutes of the preceding spectral coefficients for use in determination of the context class.

31. The method of claim 30, wherein a sum of the quantized absolutes of the preceding spectral coefficients is used for in determination of the context class

32. The method of claim 30, wherein processing the preceding spectral coefficients comprises a first quantizing step in which the absolutes of the preceding spectral coefficients are quantized according a first quantization scheme, a step of determining a variance of the absolutes of the preceding spectral coefficients quantized according the first quantization scheme, a step of using the determined variance for selecting one of at least two different non-linear second quantization schemes, and a second quantizing step in which the absolutes of the preceding spectral coefficients quantized according the first quantization scheme are further quantized according to the selected non-linear second quantization scheme.

33. The method of claim 32, wherein the step of determining the variance of the absolutes of the preceding spectral coefficients quantized according the first quantization scheme comprises determining a sum of the absolutes of the preceding spectral coefficients quantized according the first quantization scheme and comparing the determined sum with at least one threshold.

34. The method of claim 30, wherein processing either results in a first outcome or at least a different second outcome and determining the context class further comprises determining a number of preceding spectral coefficients being non-uniformly quantized to the first outcome, and using the determined number for determining the context class.

35. The method of claim 34 wherein one of the preceding spectral coefficients is preferred over the preceding spectral coefficients remaining, said method further comprises using the non-uniformly quantization of the preferred one among the preceding spectral coefficient for determining the context class.

36. The method of claim 35, wherein said preferred among the preceding spectral coefficients is comprised in a preceding spectrum and said current spectral coefficient is comprised in a different current spectrum, said preferred preceding and said current spectral coefficients being comprised at a same frequency in the respective spectrum.

37. The method of claim 36, wherein said different current spectrum further comprises at least one of the remaining preceding spectral coefficients at a frequency lower than said same frequency.

38. The method of claim 30, said method further comprises receiving at least one of a mode switching signal and a reset signal and using the at least one received signal for controlling the step of determining the context class.

39. The method of claim 30, said method further comprises using a representative set of data for determining the at least two different probability density functions.

40. The method of claim 30, wherein the mapping is realized using a look-up table or a hash table