公开(公告)号：US20230061045A1

公开(公告)日：2023-03-02

申请号：US17760076

申请日：2021-04-27

Applicant: Google LLC

Inventor： Jyrki Antero Alakuijala ,  Moritz Firsching

IPC: G06T5/00 ,  G06T5/20 ,  G06F3/14 ,  G06T5/50

Abstract: A technique for improving progressive encoded JPEG includes displaying an oversmoothed version of an image as the image data is being received. The oversmoothed image may be smoothed according to a smoothing kernel, e.g., a convolution kernel (such as a Gaussian). The oversmoothed image is a first layer over which other image layers are displayed. It is noted that the oversmoothed image may present a recognizable version of the image to a user, including recognizable versions of various image features (e.g., persons, objects). As the other layers are rendered on the display, these image features remain visible to the user. That is, the image features are not artifacts that may disappear with the rendering of final image layers; this may occur with the conventional progressive encoded images and interferes with the user experience.

公开(公告)号：US20250139831A1

公开(公告)日：2025-05-01

申请号：US18693704

申请日：2023-01-27

Applicant: GOOGLE LLC

Inventor： Jyrki Antero Alakuijala ,  Matthew Sharifi ,  Zoltan Szabadka ,  Moritz Firsching ,  Thomas Fischbacher ,  Sami Boukortt ,  Martin Bruse ,  Evgenii Kliuchnikov

IPC: G06T9/00

Abstract: A method including generating base values and delta values based on an image, generating weighted delta values based on the delta values, generating an enhanced image based on the base values and the weighted delta values, and compressing the enhanced image.

公开(公告)号：US20240205603A1

公开(公告)日：2024-06-20

申请号：US18538452

申请日：2023-12-13

Applicant: Google LLC

Inventor： Jyrki Antero Alakuijala ,  Matthew Sharifi ,  Martin Bruse ,  Zoltan Szabadka ,  Thomas Fischbacher ,  Sami Boukortt ,  Moritz Firsching ,  Evgenii Kliuchnikov

IPC: H04R3/12 ,  H03F3/217 ,  H04R1/40

CPC classification number: H04R3/12 ,  H03F3/2171 ,  H04R1/403 ,  H03F2200/03 ,  H03F2200/351 ,  H04R2201/401 ,  H04R2201/403 ,  H04R2430/20

Abstract: Spatial audio may be generated by a speaker array that is switched according to rows and/or columns to reduce its cost and complexity. The speaker array may include a row of speakers that are each coupled to a different column channel. The rows of speakers can receive portions of the spatial audio on a row-by-row basis as each row is activated to couple the speakers in a row to their respective column. This switched approach reduces a number of required audio sources. The audio sources may generate PWM signals for each column using an approach that is similar to that found in Class-D amplification or sigma-delta Modulation. Analog signals may be recovered from the PWM signals using a low-pass filter positioned before each speaker in the array.

公开(公告)号：US12223627B2

公开(公告)日：2025-02-11

申请号：US17760076

申请日：2021-04-27

Applicant: Google LLC

Inventor： Jyrki Antero Alakuijala ,  Moritz Firsching

IPC: G06T5/70 ,  G06F3/14 ,  G06T5/20 ,  G06T5/50

Abstract: A technique for improving progressive encoded JPEG includes displaying an oversmoothed version of an image as the image data is being received. The oversmoothed image may be smoothed according to a smoothing kernel, e.g., a convolution kernel (such as a Gaussian). The oversmoothed image is a first layer over which other image layers are displayed. It is noted that the oversmoothed image may present a recognizable version of the image to a user, including recognizable versions of various image features (e.g., persons, objects). As the other layers are rendered on the display, these image features remain visible to the user. That is, the image features are not artifacts that may disappear with the rendering of final image layers; this may occur with the conventional progressive encoded images and interferes with the user experience.

公开(公告)号：US20240105190A1

公开(公告)日：2024-03-28

申请号：US18472841

申请日：2023-09-22

Applicant: GOOGLE LLC

Inventor： Martin Bruse ,  Jyrki Antero Alakuijala ,  Moritz Firsching ,  Thomas Fischbacher ,  Sami Boukortt ,  Evgenii Kliuchnikov

IPC: G10L19/022

CPC classification number: G10L19/022

Abstract: A method including receiving an audio signal, generating a transformed audio signal by transforming the audio signal using a plurality of windows each separated in time, generating an interpolated audio signal by interpolating the transformed audio signal, generating a separated audio signal by applying a mask to the interpolated audio signal, and compressing the separated audio signal.

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.