公开(公告)号：US20220198717A1

公开(公告)日：2022-06-23

申请号：US17654529

申请日：2022-03-11

Applicant: Adobe Inc.

Inventor： Meet Patel ,  Mayur Hemani ,  Karanjeet Singh ,  Amit Gupta ,  Apoorva Gupta ,  Balaji Krishnamurthy

IPC: G06T9/00 ,  G06N3/08 ,  G06T7/00

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for utilizing deep learning to intelligently determine compression settings for compressing a digital image. For instance, the disclosed system utilizes a neural network to generate predicted perceptual quality values for compression settings on a compression quality scale. The disclosed system fits the predicted compression distortions to a perceptual distortion characteristic curve for interpolating predicted perceptual quality values across the compression settings on the compression quality scale. Additionally, the disclosed system then performs a search over the predicted perceptual quality values for the compression settings along the compression quality scale to select a compression setting based on a perceptual quality threshold. The disclosed system generates a compressed digital image according to compression parameters for the selected compression setting.

公开(公告)号：US11645786B2

公开(公告)日：2023-05-09

申请号：US17654529

申请日：2022-03-11

Applicant: Adobe Inc.

Inventor： Meet Patel ,  Mayur Hemani ,  Karanjeet Singh ,  Amit Gupta ,  Apoorva Gupta ,  Balaji Krishnamurthy

IPC: G06K9/00 ,  G06T9/00 ,  G06T7/00 ,  G06N3/08

CPC classification number: G06T9/002 ,  G06N3/08 ,  G06T7/0002 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/20224

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for utilizing deep learning to intelligently determine compression settings for compressing a digital image. For instance, the disclosed system utilizes a neural network to generate predicted perceptual quality values for compression settings on a compression quality scale. The disclosed system fits the predicted compression distortions to a perceptual distortion characteristic curve for interpolating predicted perceptual quality values across the compression settings on the compression quality scale. Additionally, the disclosed system then performs a search over the predicted perceptual quality values for the compression settings along the compression quality scale to select a compression setting based on a perceptual quality threshold. The disclosed system generates a compressed digital image according to compression parameters for the selected compression setting.

公开(公告)号：US11335033B2

公开(公告)日：2022-05-17

申请号：US17032704

申请日：2020-09-25

Applicant: Adobe Inc.

Inventor： Meet Patel ,  Mayur Hemani ,  Karanjeet Singh ,  Amit Gupta ,  Apoorva Gupta ,  Balaji Krishnamurthy

IPC: G06K9/00 ,  G06T9/00 ,  G06T7/00 ,  G06N3/08

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for utilizing deep learning to intelligently determine compression settings for compressing a digital image. For instance, the disclosed system utilizes a neural network to generate predicted perceptual quality values for compression settings on a compression quality scale. The disclosed system fits the predicted compression distortions to a perceptual distortion characteristic curve for interpolating predicted perceptual quality values across the compression settings on the compression quality scale. Additionally, the disclosed system then performs a search over the predicted perceptual quality values for the compression settings along the compression quality scale to select a compression setting based on a perceptual quality threshold. The disclosed system generates a compressed digital image according to compression parameters for the selected compression setting.

公开(公告)号：US20220101564A1

公开(公告)日：2022-03-31

申请号：US17032704

申请日：2020-09-25

Applicant: Adobe Inc.

Inventor： Meet Patel ,  Mayur Hemani ,  Karanjeet Singh ,  Amit Gupta ,  Apoorva Gupta ,  Balaji Krishnamurthy

IPC: G06T9/00 ,  G06T7/00 ,  G06N3/08

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for utilizing deep learning to intelligently determine compression settings for compressing a digital image. For instance, the disclosed system utilizes a neural network to generate predicted perceptual quality values for compression settings on a compression quality scale. The disclosed system fits the predicted compression distortions to a perceptual distortion characteristic curve for interpolating predicted perceptual quality values across the compression settings on the compression quality scale. Additionally, the disclosed system then performs a search over the predicted perceptual quality values for the compression settings along the compression quality scale to select a compression setting based on a perceptual quality threshold. The disclosed system generates a compressed digital image according to compression parameters for the selected compression setting.