IP Library › Granted Patent US 9,479,771
Granted Patent B2
US 9,479,771 · App. 14/155,500 · Granted Oct 25, 2016

Using multiple transform/subband representations for video encoding

Inventor: Jae S. Lim (Winchester, MA)
Assignee: MASSACHUSETTS INSTITUTE OF TECHNOLOGY
H04N19/00018H04N19/103H04N19/18H04N19/60H04N19/61H04N19/63
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,479,771
App. No.
14/155,500
Granted
Oct 25, 2016
Kind
B2
Abstract

A method of encoding a sequence of video frames, comprising receiving a sequence of video frames, encoding at least a first portion of the sequence of frames using the first type of transform/subband representation to produce first transform/subband coefficients, encoding at least a second portion of the sequence of frames using the second type of transform/subband representation to produce second transform/subband coefficients; and providing the first and second transform/subband coefficients at an output.

Claims (112)

1. A method of encoding a sequence of video frames using an encoder, comprising:

receiving a sequence of video frames;

encoding at least a first portion of the sequence of frames using a first type of transform/subband representation to produce first transform/subband coefficients;

encoding at least a second portion of the sequence of frames, different from the first portion of the sequence of frames, using a second type of transform/subband representation to produce second transform/subband coefficients, wherein the second type of transform/subband representation is different from the first type of transform/subband representation, and wherein the first type and second type of transform/subband representation are used for encoding within a first region and within a second region, respectively, of the same frame, so that the type of transform/subband representation differs from region to region within the same frame; and

providing the first and second transform/subband coefficients at an output.

2. A method of decoding a sequence of encoded video frames using a decoder, comprising:

receiving a sequence of first and second transform/subband coefficients representing a sequence of video frames;

decoding the first transform/subband coefficients using a first type of transform/subband representation to produce first portions of the sequence of video frames;

decoding the second transform/subband coefficients using a second type of transform/subband representation to produce second portions of the sequence of video frames, different from the first portions of the sequence of video frames, wherein the second type of transform/subband representation is different from the first type of transform/subband representation, and wherein the first type and second type of transform/subband representation are used for decoding within a first region and within a second region, respectively, of the same frame, so that the type of transform/subband representation differs from region to region within the same frame; and

providing the first and second portions of the sequence of video frames at an output.

3. A communication system receiver for decoding a sequence of encoded video frames, comprising:

processing circuitry configured to receive a sequence of first and second transform/subband coefficients representing a sequence of video frames;

processing circuitry configured to decode the first transform/subband coefficients using a first type of transform/subband representation to produce first portions of the sequence of video frames;

processing circuitry configured to decode the second transform/subband coefficients using a second type of transform/subband representation to produce second portions of the sequence of video frames, different from the first portions of the sequence of video frames, wherein the second type of transform/subband representation is different from the first type of transform/subband representation, and wherein the first type and second type of transform/subband representation are used for decoding within a first region and within a second region, respectively, of the same frame, so that the type of transform/subband representation differs from region to region within the same frame; and

processing circuitry configured to provide the first and second portions of the sequence of video frames at an output.

4. The method of claim 1 wherein the first type and second type of transform/subband representation are used for encoding a first and a second frame, respectively, so that the type of transform/subband representation differs from frame to frame.

5. The method of claim 4 wherein the first type of transform/subband representation is used for encoding I-frames in a communication system, and the second type of transform/subband representation is used for encoding B and P-frames in the communication system.

6. The method of claim 1 wherein the first region has primarily inter-frame coded blocks and the second region has primarily intra-frame coded blocks.

7. The method of claim 1 wherein the first and second regions both contain only inter-frame coded blocks.

8. The method of claim 1 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is a Non-ORT.

9. The method of claim 1 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is an ORT.

10. The method of claim 9 wherein the Non-ORT is a DCT.

11. The method of claim 1 further comprising the step of quantizing the transform/subband coefficients.

12. The method of claim 1 further comprising providing identification information along with the transform/subband coefficients to identify at least one of the first and second type of transform/subband representations.

13. The method of claim 1 wherein identification information to identify the type of transform/subband representation used for the first or second portion of the sequence is not supplied with the transform/subband coefficients, as the identity of the representation will be known to a decoder.

14. The method of claim 1 wherein at least a third portion of the sequence is encoded with a third type of transform/subband representation different from the first and second types.

15. The method of claim 1 further comprising analyzing at least the first portion by each of multiple different transform/subband representations, comparing results of the analysis, and selecting the first type of transform/subband representation based on the comparison for encoding the first portion.

16. The method of claim 2 wherein the first type and second type of transform/subband representation are used for decoding a first and a second frame, respectively, so that the type of transform/subband representation differs from frame to frame.

17. The method of claim 16 wherein the first type of transform/subband representation is used for decoding I-frames in a communication system, and the second type of transform/subband representation is used for decoding B and P-frames in the communication system.

18. The method of claim 2 wherein the first region has primarily inter-frame coded blocks and the second region has primarily intra-frame coded blocks.

19. The method of claim 2 wherein the first and second regions both contain only inter-frame coded blocks.

20. The method of claim 2 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is a Non-ORT.

21. The method of claim 2 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is an ORT.

22. The method of claim 21 wherein the Non-ORT is a DCT.

23. The method of claim 2 further comprising the step of quantizing the transform/subband coefficients.

24. The method of claim 2 further comprising providing identification information along with the transform/subband coefficients to identify at least one of the first and second type of transform/subband representations.

25. The method of claim 2 wherein identification information to identify the type of transform/subband representation used for the first or second portion of the sequence is not supplied with the transform/subband coefficients, as the identity of the representation will be known to a decoder.

26. The method of claim 2 wherein at least a third portion of the sequence is encoded with a third type of transform/subband representation different from the first and second types.

27. The receiver of claim 3 wherein the first type and second type of transform/subband representation are used for decoding a first and a second frame, respectively, so that the type of transform/subband representation differs from frame to frame.

28. The receiver of claim 27 wherein the first type of transform/subband representation is used for decoding I-frames in a communication system, and the second type of transform/subband representation is used for decoding B and P-frames in the communication system.

29. The receiver of claim 3 wherein the first region has primarily inter-frame coded blocks and the second region has primarily intra-frame coded blocks.

30. The receiver of claim 3 wherein the first and second regions both contain only inter-frame coded blocks.

31. The receiver of claim 3 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is a Non-ORT.

32. The receiver of claim 3 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is an ORT.

33. The receiver of claim 32 wherein the Non-ORT is a DCT.

34. The receiver of claim 3 further comprising the step of quantizing the transform/subband coefficients.

35. The receiver of claim 3 further comprising providing identification information along with the transform/subband coefficients to identify at least one of the first and second type of transform/subband representations.

36. The receiver of claim 3 wherein identification information to identify the type of transform/subband representation used for the first or second portion of the sequence is not supplied with the transform/subband coefficients, as the identity of the representation will be known to a decoder.

37. The receiver of claim 3 wherein at least a third portion of the signal is encoded with a third type of transform/subband representation different from the first and second types.

38. The method of claim 8 wherein the first type of transform/subband representation is a DCT, and the second type of transform/subband representation is a DFT.

39. The method of claim 1 wherein the first type of transform/subband representation and the second type of transform/subband representation are used for encoding a B-frame.

40. The method of claim 39 wherein the first type of transform/subband representation is a DCT.

41. The method of claim 40 wherein the second type of transform/subband representation is a DFT.

42. The method of claim 39 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is a Non-ORT.

43. The method of claim 42 wherein the first type of transform/subband representation is a DCT.

44. The method of claim 43 wherein the second type of transform/subband representation is a DFT.

45. The method of claim 39 further comprising providing identification information along with the transform/subband coefficients to identify at least one of the first and second type of transform/subband representations.

46. The method of claim 39 wherein identification information to identify the type of transform/subband representation used for the first or second portion of the sequence is not supplied with the transform/subband coefficients, as the identity of the representation will be known to a decoder.

47. The method of claim 1 wherein the first type of transform/subband representation and the second type of transform/subband representation are used for encoding a P-frame.

48. The method of claim 47 wherein the first type of transform/subband representation is a DCT.

49. The method of claim 48 wherein the second type of transform/subband representation is a DFT.

50. The method of claim 47 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is a Non-ORT.

51. The method of claim 50 wherein the first type of transform/subband representation is a DCT.

52. The method of claim 51 wherein the second type of transform/subband representation is a DFT.

53. The method of claim 1 wherein the first type of transform/subband representation and the second type of transform/subband representation are used for encoding intra-frame coded blocks.

54. The method of claim 53 wherein the first type of transform/subband representation is a DCT.

55. The method of claim 54 wherein the second type of transform/subband representation is a DFT.

56. The method of claim 53 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is a Non-ORT.

57. The method of claim 56 wherein the first type of transform/subband representation is a DCT.

58. The method of claim 57 wherein the second type of transform/subband representation is a DFT.

59. The method of claim 20 wherein the first type of transform/subband representation is a DCT, and the second type of transform/subband representation is a DFT.

60. The method of claim 2 wherein the first type of transform/subband representation and the second type of transform/subband representation are used for decoding a B-frame.

61. The method of claim 60 wherein the first type of transform/subband representation is a DCT.

62. The method of claim 61 wherein the second type of transform/subband representation is a DFT.

63. The method of claim 60 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is a Non-ORT.

64. The method of claim 63 wherein the first type of transform/subband representation is a DCT.

65. The method of claim 64 wherein the second type of transform/subband representation is a DFT.

66. The method of claim 60 further comprising providing identification information along with the transform/subband coefficients to identify at least one of the first and second type of transform/subband representations.

67. The method of claim 60 wherein identification information to identify the type of transform/subband representation used for the first or second portion of the sequence is not supplied with the transform/subband coefficients, as the identity of the representation will be known to a decoder.

68. The method of claim 2 wherein the first type of transform/subband representation and the second type of transform/subband representation are used for decoding a P-frame.

69. The method of claim 68 wherein the first type of transform/subband representation is a DCT.

70. The method of claim 69 wherein the second type of transform/subband representation is a DFT.

71. The method of claim 68 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is a Non-ORT.

72. The method of claim 71 wherein the first type of transform/subband representation is a DCT.

73. The method of claim 72 wherein the second type of transform/subband representation is a DFT.

74. The method of claim 2 wherein the first type of transform/subband representation and the second type of transform/subband representation are used for decoding intra-frame coded blocks.

75. The method of claim 74 wherein the first type of transform/subband representation is a DCT.

76. The method of claim 75 wherein the second type of transform/subband representation is a DFT.

77. The method of claim 74 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is a Non-ORT.

78. The method of claim 77 wherein the first type of transform/subband representation is a DCT.

79. The method of claim 78 wherein the second type of transform/subband representation is a DFT.

80. The receiver of claim 31 wherein the first type of transform/subband representation is a DCT, and the second type of transform/subband representation is a DFT.

81. The receiver of claim 3 wherein the first type of transform/subband representation and the second type of transform/subband representation are used for decoding a B-frame.

82. The receiver of claim 81 wherein the first type of transform/subband representation is a DCT.

83. The receiver of claim 82 wherein the second type of transform/subband representation is a DFT.

84. The receiver of claim 83 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is a Non-ORT.

85. The receiver of claim 84 wherein the first type of transform/subband representation is a DCT.

86. The receiver of claim 85 wherein the second type of transform/subband representation is a DFT.

87. The receiver of claim 81 further comprising providing identification information along with the transform/subband coefficients to identify at least one of the first and second type of transform/subband representations.

88. The receiver of claim 81 wherein identification information to identify the type of transform/subband representation used for the first or second portion of the sequence is not supplied with the transform/subband coefficients, as the identity of the representation will be known to a decoder.

89. The receiver of claim 3 wherein the first type of transform/subband representation and the second type of transform/subband representation are used for decoding a P-frame.

90. The receiver of claim 89 wherein the first type of transform/subband representation is a DCT.

91. The receiver of claim 90 wherein the second type of transform/subband representation is a DFT.

92. The receiver of claim 89 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is a Non-ORT.

93. The receiver of claim 92 wherein the first type of transform/subband representation is a DCT.

94. The receiver of claim 93 wherein the second type of transform/subband representation is a DFT.

95. The receiver of claim 3 wherein the first type of transform/subband representation and the second type of transform/subband representation are used for decoding intra-frame coded blocks.

96. The receiver of claim 95 wherein the first type of transform/subband representation is a DCT.

97. The receiver of claim 96 wherein the second type of transform/subband representation is a DFT.

98. The receiver of claim 95 wherein the first type of transform/subband representation is a Non-ORT and the second type of transform/subband representation is a Non-ORT.

99. The receiver of claim 98 wherein the first type of transform/subband representation is a DCT.

100. The receiver of claim 99 wherein the second type of transform/subband representation is a DFT.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 15, 2014
From: LIM, JAE S.
To: MASSACHUSETTS INSTITUTE OF TECHNOLOGY
Reel/Frame 031971/0755 →
Continuity (2)
Continuation 11430601 · May 9, 2006
Related Publication 20140126650A1 · May 8, 2014