公开(公告)号：US20240320808A1

公开(公告)日：2024-09-26

申请号：US18734000

申请日：2024-06-05

Applicant: Google LLC

Inventor： Yicheng Wu ,  Qiurui He ,  Tianfan Xue ,  Rahul Garg ,  Jiawen Chen ,  Jonathan T. Barron

IPC: G06T5/80 ,  G06T3/40 ,  G06T5/10 ,  G06T5/20 ,  G06T7/80

CPC classification number: G06T5/80 ,  G06T3/40 ,  G06T5/10 ,  G06T5/20 ,  G06T7/80 ,  G06T2207/20081 ,  G06T2207/20084

Abstract: A method includes obtaining an input image that contains a particular representation of lens flare, and processing the input image by a machine learning model to generate a de-flared image that includes the input image with at least part of the particular representation of lens flare removed. The machine learning (ML) model may be trained by generating training images that combine respective baseline images with corresponding lens flare images. For each respective training image, a modified image may be determined by processing the respective training image by the ML model, and a loss value may be determined based on a loss function comparing the modified image to a corresponding baseline image used to generate the respective training image. Parameters of the ML model may be adjusted based on the loss value determined for each respective training image and the loss function.

公开(公告)号：US20220375045A1

公开(公告)日：2022-11-24

申请号：US17625994

申请日：2020-11-09

Applicant: Google LLC

Inventor： Yicheng Wu ,  Qiurui He ,  Tianfan Xue ,  Rahul Garg ,  Jiawen Chen ,  Jonathan T. Barron

IPC: G06T5/00 ,  G06T5/20 ,  G06T7/80 ,  G06T5/10 ,  G06T3/40

Abstract: A method includes obtaining an input image that contains a particular representation of lens flare, and processing the input image by a machine learning model to generate a de-flared image that includes the input image with at least part of the particular representation of lens flare removed. The machine learning (ML) model may be trained by generating training images that combine respective baseline images with corresponding lens flare images. For each respective training image, a modified image may be determined by processing the respective training image by the ML model, and a loss value may be determined based on a loss function comparing the modified image to a corresponding baseline image used to generate the respective training image. Parameters of the ML model may be adjusted based on the loss value determined for each respective training image and the loss function.

公开(公告)号：US12033309B2

公开(公告)日：2024-07-09

申请号：US17625994

申请日：2020-11-09

Applicant: Google LLC

Inventor： Yicheng Wu ,  Qiurui He ,  Tianfan Xue ,  Rahul Garg ,  Jiawen Chen ,  Jonathan T. Barron

IPC: G06T7/80 ,  G06T3/40 ,  G06T5/10 ,  G06T5/20 ,  G06T5/80

CPC classification number: G06T5/80 ,  G06T3/40 ,  G06T5/10 ,  G06T5/20 ,  G06T7/80 ,  G06T2207/20081 ,  G06T2207/20084

Abstract: A method includes obtaining an input image that contains a particular representation of lens flare, and processing the input image by a machine learning model to generate a de-flared image that includes the input image with at least part of the particular representation of lens flare removed. The machine learning (ML) model may be trained by generating training images that combine respective baseline images with corresponding lens flare images. For each respective training image, a modified image may be determined by processing the respective training image by the ML model, and a loss value may be determined based on a loss function comparing the modified image to a corresponding baseline image used to generate the respective training image. Parameters of the ML model may be adjusted based on the loss value determined for each respective training image and the loss function.