公开(公告)号：US20210014532A1

公开(公告)日：2021-01-14

申请号：US16707736

申请日：2019-12-09

Applicant: Google LLC

Inventor： Jyrki Antero Alakuijala ,  Luca Versari

IPC: H04N19/61 ,  H04N19/176 ,  H04N19/172 ,  H04N19/107 ,  H04N19/159 ,  H04N19/124 ,  H04N19/137

Abstract: A method can include compressing a first original frame of a video stream to an intraframe, the intraframe comprising fewer symbols than the first original frame, compressing a second original frame to a first interframe, the first interframe referencing the intraframe and comprising fewer symbols than the second original frame, determining an intraframe error of the intraframe due to the compression of the first original frame, determining a first interframe error of the first interframe due to the compression of the second original frame, determining a compression level for a third original frame based on the intraframe error and the first interframe error, and compressing the third original frame to a second interframe, the second interframe referencing the intraframe and the first interframe and comprising fewer symbols than the third original frame, a number of symbols included in the second interframe being based on the determined compression level.

公开(公告)号：US12236554B2

公开(公告)日：2025-02-25

申请号：US17766897

申请日：2019-10-14

Applicant: GOOGLE LLC

Inventor： Jyrki Alakuijala ,  Luca Versari

IPC: G06V10/00 ,  G06T3/40 ,  G06T9/00

Abstract: Method are provided that exhibit increased quality and compression factor for compressing images. The methods can include generating a set of coefficients indicative of image contents of a block of image pixels at a plurality of spatial frequencies. The set of coefficients is scaled to generate a first set of scaled coefficients. An assessment is performed for a plurality of quantization levels, which includes quantizing a subset of the first set of scaled coefficients according to respective quantization levels to generate a quantized subset of the first set of scaled coefficients and determining a post-quantization energy of the quantized subset of the first set of scaled coefficients. Based on the assessment of the plurality of quantization levels, a scaled and quantized version of the set of coefficients is generated. An encoded version of the image based on the scaled and quantized version of the set of coefficients is generated.

公开(公告)号：US20230115065A1

公开(公告)日：2023-04-13

申请号：US17959962

申请日：2022-10-04

Applicant: Google LLC

Inventor： Thomas Fischbacher ,  Luca Versari

IPC: G06N10/80

Abstract: Implementations disclosed describe techniques used for compiling a quantum algorithm for execution on a plurality of quantum circuits, including accessing, by a processing device, the quantum algorithm, identifying a matrix associated with the quantum algorithm, determining a representation of the identified matrix as a matrix decomposition that includes a plurality of transformation matrices, wherein one or more of the plurality of transformation matrices perform multiple instances of two-dimensional rotations; and generating a circuit map that maps execution of the matrix decomposition on the plurality of quantum circuits.

公开(公告)号：US20230141888A1

公开(公告)日：2023-05-11

申请号：US17917540

申请日：2020-04-08

Applicant: Google LLC

Inventor： Jyrki Alakuijala ,  Luca Versari

IPC: H04N19/119 ,  H04N19/154 ,  H04N19/86 ,  H04N19/176

CPC classification number: H04N19/119 ,  H04N19/154 ,  H04N19/86 ,  H04N19/176

Abstract: A method for partitioning a block of an image to reduce quantization artifacts includes estimating an expected entropy of the block; partitioning the block into sub-blocks, where each sub-block having a size of a smallest possible partition size; calculating respective amounts of visual masking for the sub-blocks; selecting, as a visual masking characteristic of the block, a highest visual masking value of the respective amounts of visual masking for the sub-blocks; combining the visual masking characteristic of the block and the expected entropy of the block to obtain a splitting indicator value; and determining whether to split the block based on the splitting indicator.
The changes to the abstract are shown below:
A method for partitioning a block of an image to reduce quantization artifacts includes estimating an expected entropy of the block; partitioning the block into sub-blocks, where each sub-block having a size of a smallest possible partition size; calculating respective amounts of visual masking for the sub-blocks; selecting, as a visual masking characteristic of the block, a highest visual masking value of the respective amounts of visual masking for the sub-blocks; combining the visual masking characteristic of the block and the expected entropy of the block to obtain a splitting indicator value; and determining whether to split the block based on the splitting indicator.

公开(公告)号：US11228786B2

公开(公告)日：2022-01-18

申请号：US16707736

申请日：2019-12-09

Applicant: Google LLC

Inventor： Jyrki Antero Alakuijala ,  Luca Versari

IPC: H04N19/137 ,  H04N19/159 ,  H04N19/172 ,  H04N19/124 ,  H04N19/176 ,  H04N19/61 ,  H04N19/107

Abstract: A method can include compressing a first original frame of a video stream to an intraframe, the intraframe comprising fewer symbols than the first original frame, compressing a second original frame to a first interframe, the first interframe referencing the intraframe and comprising fewer symbols than the second original frame, determining an intraframe error of the intraframe due to the compression of the first original frame, determining a first interframe error of the first interframe due to the compression of the second original frame, determining a compression level for a third original frame based on the intraframe error and the first interframe error, and compressing the third original frame to a second interframe, the second interframe referencing the intraframe and the first interframe and comprising fewer symbols than the third original frame, a number of symbols included in the second interframe being based on the determined compression level.

公开(公告)号：US20230147376A1

公开(公告)日：2023-05-11

申请号：US17766897

申请日：2019-10-14

Applicant: GOOGLE LLC

Inventor： Jyrki Alakuijala ,  Luca Versari

IPC: G06T3/40 ,  G06T9/00

CPC classification number: G06T3/40 ,  G06T9/00

Abstract: Methods are provided for improving the quality and compression factor of compressed images. The methods include determining frequency-band-specific quantization levels on a block-by-block level. This results in an adaptive dead zone, allowing certain blocks to be represented by fewer nonzero elements while other blocks are represented by more nonzero elements. Accordingly, the quality of the encoded image is improved while maintaining or improving the compression ratio. The adaptive quantization level is determined by comparing a post-quantization energy level to a threshold energy criterion for each frequency band within a block. Where the energy threshold criterion is not satisfied via these methods, additional methods can be applied to improve the image quality. The methods described herein allow the effective compression ratio of the image to be adapted on a block-by-block, frequency-sensitive basis in order to more effectively allocate encoded image bits where they will have the most effect on image quality.

公开(公告)号：US20210256388A1

公开(公告)日：2021-08-19

申请号：US17169740

申请日：2021-02-08

Applicant: Google LLC

Inventor： Thomas Fischbacher ,  Luca Versari ,  Krzysztof Potempa ,  Iulia-Maria Comsa ,  Moritz Firsching ,  Jyrki Antero Alakuijala

IPC: G06N3/08 ,  G06F17/16 ,  G06N3/04

Abstract: The present disclosure proposes a model that has more expressive power, e.g., can generalize from a smaller amount of parameters and assign more computation in areas of the function that need more computation. In particular, the present disclosure is directed to novel machine learning architectures that use the exponential of an input-dependent matrix as a nonlinearity. The mathematical simplicity of this architecture allows a detailed analysis of its behavior.

公开(公告)号：US20210248476A1

公开(公告)日：2021-08-12

申请号：US17170025

申请日：2021-02-08

Applicant: Google LLC

Inventor： Thomas Fischbacher ,  Iulia-Maria Comsa ,  Luca Versari

IPC: G06N3/08 ,  G06N3/04 ,  G06N20/00 ,  G06K9/62 ,  G06F17/16

Abstract: The present disclosure proposes a model that has more expressive power, e.g., can generalize from a smaller amount of parameters and assign more computation in areas of the function that need more computation. In particular, the present disclosure is directed to novel machine learning architectures that use the exponential of an input-dependent matrix as a nonlinearity. The mathematical simplicity of this architecture allows a detailed analysis of its behavior, providing stringent robustness guarantees via Lipschitz bounds.