US-12627838-B2 - Block-based predictive coding and decoding of a picture

US12627838B2US 12627838 B2US12627838 B2US 12627838B2US-12627838-B2

Abstract

A previously encoded or reconstructed version of a neighborhood of a predetermined block to be predicted is exploited so as to result into a more efficient predictive coding of the prediction block. In particular, a spectral decomposition of a region composed of this neighborhood and a first version of a predicted filling of the predetermined block results into a first spectrum which is subject to noise reduction and the thus resulting second spectrum may be subject to a spectral composition, thereby resulting in a modified version of this region including a second version of the predicted filling of the predetermined block. Owing to the exploitation of the already processed, i.e. encoded/reconstructed, neighborhood of the predetermined block, the second version of the predicted filling of the predetermined block tends to improve the coding efficiency.

Inventors

Jonathan PFAFF
Martin Winken
Christian RUDAT
Heiko Schwarz
Thomas Wiegand
Detlev Marpe

Assignees

Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V.

Dates

Publication Date: 20260512
Application Date: 20240827
Priority Date: 20170105

Claims (20)

1 . A method for decoding a picture from a data stream, the method comprising: deriving, from the data stream, a quantization parameter associated with a block of the picture; applying a transform to a region of the picture, the region including a first set of samples included in the block and a second set of samples that neighbor the block; determining a threshold based at least in part on the quantization parameter and a lookup table; filtering the transformed region using the threshold; applying an inverse of the transform to the filtered region to obtain a modified version of the region; and reconstructing the block using samples included in the modified version of the region that corresponds to respective positions of the first set of samples.
2 . The method of claim 1 , wherein filtering the transformed region comprises: comparing an absolute value of a sample within the transformed region to the determined threshold; and when the absolute value of the sample is less than the determined threshold, modifying a value of the sample.
3 . The method of claim 1 , further comprising: identifying a set of thresholds, from multiple set of thresholds, for noise reduction based on the quantization parameter; decoding a noise reduction signal; and selecting the threshold from the identified set of thresholds based on the noise reduction signal.
4 . The method of claim 3 , wherein each of the sets of thresholds includes eight thresholds.
5 . The method of claim 1 , wherein: the first set of samples represent a prediction signal of the block; and the second set of samples represent a reconstructed portion of the picture that neighbors the block.
6 . The method of claim 5 , wherein: the reconstructed portion of the picture is L-shaped and includes a lower portion and a right portion; the lower portion is co-linear with a left side of the block along a vertical axis; and the right portion is co-linear with a top side of the block along a horizontal axis.
7 . The method of claim 1 , further comprising: determining whether to reconstruct the block using a prediction residual and the first set of samples included in the block; in response to a first determination, reconstructing the block using the prediction residual and the first set of samples included in the block; and in response to a second determination, reconstructing the block using the prediction residual and a portion of the samples included in the modified version of the region.
8 . An electronic device for decoding a picture from a data stream, the electronic device comprising: a processor configured to: derive, from the data stream, a quantization parameter associated with a block of the picture; apply a transform to a region of the picture, the region including a first set of samples included in the block and a second set of samples that neighbor the block; determine a threshold based at least in part on the quantization parameter and a lookup table; filter the transformed region using the threshold; apply an inverse of the transform to the filtered region to obtain a modified version of the region; and reconstruct the block using samples included in the modified version of the region that corresponds to respective positions of the first set of samples.
9 . The electronic device of claim 8 , wherein to filter the transformed region the processor is further configured to: compare an absolute value of a sample within the transformed region to the determined threshold; and when the absolute value of the sample is less than the determined threshold, modify a value of the sample.
10 . The electronic device of claim 8 , wherein the processor is further configured to: identify a set of thresholds, from multiple set of thresholds, for noise reduction based on the quantization parameter; decode a noise reduction signal; and select the threshold from the identified set of thresholds based on the noise reduction signal.
11 . The electronic device of claim 10 , wherein each of the sets of thresholds includes eight thresholds.
12 . The electronic device of claim 8 , wherein: the first set of samples represent a prediction signal of the block; and the second set of samples represent a reconstructed portion of the picture that neighbors the block.
13 . The electronic device of claim 12 , wherein: the reconstructed portion of the picture is L-shaped and includes a lower portion and a right portion; the lower portion is co-linear with a left side of the block along a vertical axis; and the right portion is co-linear with a top side of the block along a horizontal axis.
14 . The electronic device of claim 8 , wherein the processor is further configured to: determine whether to reconstruct the block using a prediction residual and the first set of samples included in the block; in response to a first determination, reconstruct the block using the prediction residual and the first set of samples included in the block; and in response to a second determination, reconstruct the block using the prediction residual and a portion of the samples included in the modified version of the region.
15 . A non-transitory, computer-readable medium storing instructions that, when executed by at least one processor of an electronic device, cause the electronic device to: derive a quantization parameter associated with a block of a picture; apply a transform to a region of the picture, the region including a first set of samples included in the block and a second set of samples that neighbor the block; determine a threshold based at least in part on the quantization parameter and a lookup table; filter the transformed region using the threshold; apply an inverse of the transform to the filtered region to obtain a modified version of the region; and reconstruct the block using samples included in the modified version of the region that corresponds to respective positions of the first set of samples.
16 . The non-transitory, computer-readable medium of claim 15 , wherein the instructions that when executed cause the at least one processor to filter the transformed region, further comprise instructions that when executed cause the at least one processor to: compare an absolute value of a sample within the transformed region to the determined threshold; and when the absolute value of the sample is less than the determined threshold, modify a value of the sample.
17 . The non-transitory, computer-readable medium of claim 15 , further containing instructions that when executed cause the at least one processor to: identify a set of thresholds, from multiple set of thresholds, for noise reduction based on the quantization parameter, wherein each of the sets of thresholds includes eight thresholds; decode a noise reduction signal; and select the threshold from the identified set of thresholds based on the noise reduction signal.
18 . The non-transitory, computer-readable medium of claim 15 , wherein: the first set of samples represent a prediction signal of the block; and the second set of samples represent a reconstructed portion of the picture that neighbors the block.
19 . The non-transitory, computer-readable medium of claim 18 , wherein: the reconstructed portion of the picture is L-shaped and includes a lower portion and a right portion; the lower portion is co-linear with a left side of the block along a vertical axis; and the right portion is co-linear with a top side of the block along a horizontal axis.
20 . The non-transitory, computer-readable medium of claim 15 , further containing instructions that when executed cause the at least one processor to: determine whether to reconstruct the block using a prediction residual and the first set of samples included in the block; in response to a first determination, reconstruct the block using the prediction residual and the first set of samples included in the block; and in response to a second determination, reconstruct the block using the prediction residual and a portion of the samples included in the modified version of the region.

Description

CROSS-REFERENCES TO RELATED APPLICATIONS This application is a continuation of U.S. patent application Ser. No. 18/211,729 filed Jun. 20, 2023, which is a continuation of U.S. patent application Ser. No. 17/465,611 filed Sep. 2, 2021 (U.S. Pat. No. 11,729,426 issued Aug. 15, 2023) which is a continuation of U.S. patent application Ser. No. 16/503,004 filed Jul. 3, 2019 (U.S. Pat. No. 11,115,682 issued Sep. 7, 2021), which is a continuation of PCT International Patent Application No. PCT/EP2017/083789, filed Dec. 20, 2017, which claims priority from European Patent Application No. 17150454.1, filed Jan. 5, 2017, the entire contents of each of are herein incorporated by reference. The present application is concerned with block-based predictive coding and decoding of pictures such as applicable in hybrid video codecs, for example. BACKGROUND OF THE INVENTION Nowadays many video codecs and still picture codecs use block-based predictive coding to compress the data used to represent the picture content. The better the prediction is, the lower the data needed to code the prediction residual. The overall benefit from using prediction depends on the amount of data needed to keep the prediction synchronized between encoder and decoder, i.e. the data needed for prediction parameterization. SUMMARY According to an embodiment, an apparatus for block-based predictive decoding of a picture may have: a prediction provider configured to predict a predetermined block of the picture to acquire a first version of a predicted filling of the predetermined block; a spectral decomposer configured to spectrally decompose a region composed of the first version of the predicted filling of the predetermined block and a reconstructed version of a neighborhood of the predetermined block so as to acquire a first spectrum of the region; a noise reducer configured to perform noise reduction on the first spectrum to acquire a second spectrum; a spectral composer configured to subject the second spectrum to spectral composition so as to acquire a modified version of the region including a second version of the predicted filling of the predetermined block; a reconstructor configured to decode a reconstructed version of the predetermined block from a data stream on the basis of the second version of the predicted filling. According to another embodiment, an apparatus for block-based predictive encoding of a picture may have: a prediction provider configured to predict a predetermined block of the picture to acquire a first version of a predicted filling of the predetermined block; a spectral decomposer configured to spectrally decompose a region composed of the first version of the predicted filling of the predetermined block and a previously encoded version of a neighborhood of the predetermined block so as to acquire a first spectrum of the region; a noise reducer configured to perform noise reduction on the first spectrum to acquire a second spectrum; a spectral composer configured to subject the second spectrum to spectral composition so as to acquire a modified version of the region including a second version of the predicted filling of the predetermined block; an encoding stage configured to encode the predetermined block into a data stream on the basis of the second version of the predicted filling. According to another embodiment, a method for block-based predictive decoding of a picture may have the steps of: predicting a predetermined block of the picture to acquire a first version of a predicted filling of the predetermined block; spectrally decomposing a region composed of the first version of the predicted filling of the predetermined block and a reconstructed version of a neighborhood of the predetermined block so as to acquire a first spectrum of the region; performing noise reduction on the first spectrum to acquire a second spectrum; subjecting the second spectrum to spectral composition so as to acquire a modified version of the region including a second version of the predicted filling of the predetermined block; decoding a reconstructed version of the predetermined block from a data stream on the basis of the second version of the predicted filling. According to another embodiment, a method for block-based predictive encoding of a picture may have the steps of: predicting a predetermined block of the picture to acquire a first version of a predicted filling of the predetermined block; spectrally decomposing a region composed of the first version of the predicted filling of the predetermined block and a previously encoded version of a neighborhood of the predetermined block so as to acquire a first spectrum of the region; performing noise reduction on the first spectrum to acquire a second spectrum; subjecting the second spectrum to spectral composition so as to acquire a modified version of the region including a second version of the predicted filling of the predetermined block; encoding the predetermined block into a da