-
1.发明公开公开(公告)号：US20230237628A1
公开(公告)日：2023-07-27
申请号：US17582266
申请日：2022-01-24
Applicant: Adobe Inc.
Inventor： Michael Gharbi , Camille Biscarrat
CPC classification number: G06T5/50 , G06T7/337 , G06N3/08 , G06T2207/20212 , G06T2207/20084 , G06T2207/20081
Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media that utilize a continuous kernel neural network that learns continuous reconstruction kernels to merge digital image samples in local neighborhoods and generate enhanced digital images from a plurality of burst digital images. For example, the disclosed systems can utilize an alignment model to align image samples from burst digital images to a common coordinate system (e.g., without resampling). In some embodiments, the disclosed systems generate localized latent vector representations of kernel neighborhoods and determines continuous displacement vectors between the image samples and output pixels of the enhanced digital image. The disclosed systems can utilize the continuous kernel network together with the latent vector representations and continuous displacement vectors to generated learned kernel weights for combining the image samples and generating an enhanced digital image.
-
公开(公告)号：US20210158495A1
公开(公告)日：2021-05-27
申请号：US16692843
申请日：2019-11-22
Applicant: Adobe Inc.
Inventor： Connelly Barnes , Utkarsh Singhal , Elya Shechtman , Michael Gharbi
Abstract: A method for manipulating a target image includes generating a query of the target image and keys and values of a first reference image. The method also includes generating matching costs by comparing the query of the target image with each key of the reference image and generating a set of weights from the matching costs. Further, the method includes generating a set of weighted values by applying each weight of the set of weights to a corresponding value of the values of the reference image and generating a weighted patch by adding each weighted value of the set of weighted values together. Additionally, the method includes generating a combined weighted patch by combining the weighted patch with additional weighted patches associated with additional queries of the target image and generating a manipulated image by applying the combined weighted patch to an image processing algorithm.
-
公开(公告)号：US20240394834A1
公开(公告)日：2024-11-28
申请号：US18323233
申请日：2023-05-24
Applicant: Adobe Inc.
Inventor： Zhihao Xia , Michael Gharbi , Jiawen Chen , Goutam Bhat
Abstract: The present disclosure relates to systems, methods, and non-transitory computer-readable media that implements self-supervised training of an image burst model, trained exclusively on low-resolution images. For example, the disclosed system accesses an image burst that includes a plurality of images. The disclosed system generates a high-resolution image estimation from a first subset of images of the plurality of images. Further, the disclosed system generates a set of low-resolution images by modifying the high-resolution image estimation based on parameters of one or more images from the plurality of images. Moreover, the disclosed system determines a measure of loss by comparing the set of low-resolution images with a second subset of images from the plurality of images and updates the image burst model with the determined measure of loss.
-
4.发明授权公开(公告)号：US12079957B2
公开(公告)日：2024-09-03
申请号：US17582266
申请日：2022-01-24
Applicant: Adobe Inc.
Inventor： Michael Gharbi , Camille Biscarrat
CPC classification number: G06T5/50 , G06N3/08 , G06T7/337 , G06T2207/20081 , G06T2207/20084 , G06T2207/20212
Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media that utilize a continuous kernel neural network that learns continuous reconstruction kernels to merge digital image samples in local neighborhoods and generate enhanced digital images from a plurality of burst digital images. For example, the disclosed systems can utilize an alignment model to align image samples from burst digital images to a common coordinate system (e.g., without resampling). In some embodiments, the disclosed systems generate localized latent vector representations of kernel neighborhoods and determines continuous displacement vectors between the image samples and output pixels of the enhanced digital image. The disclosed systems can utilize the continuous kernel network together with the latent vector representations and continuous displacement vectors to generated learned kernel weights for combining the image samples and generating an enhanced digital image.
-
公开(公告)号：US11783184B2
公开(公告)日：2023-10-10
申请号：US17590995
申请日：2022-02-02
Applicant: Adobe Inc.
Inventor： Federico Perazzi , Zhihao Xia , Michael Gharbi , Kalyan Sunkavalli
Abstract: Certain embodiments involve techniques for efficiently estimating denoising kernels for generating denoised images. For instance, a neural network receives a noisy reference image to denoise. The neural network uses a kernel dictionary of base kernels and generates a coefficient vector for each pixel in the reference image such that the coefficient vector includes a coefficient value for each base kernel in the kernel dictionary, where the base kernels are combined to generate a denoising kernel and each coefficient value indicates a contribution of a given base kernel to a denoising kernel. The neural network calculates the denoising kernel for a given pixel by applying the coefficient vector for that pixel to the kernel dictionary. The neural network applies each denoising kernel to the respective pixel to generate a denoised output image.
-
Patent Agency Ranking
公开(公告)号：US11756264B2
公开(公告)日：2023-09-12
申请号：US17534225
申请日：2021-11-23
Applicant: Adobe Inc.
Inventor： Chi Cheng Hsu , Michal Lukác , Michael Gharbi , Kevin Wampler
CPC classification number: G06T17/20 , G06T11/203
Abstract: Embodiments are disclosed for receiving a target shape. The method may further include initializing a gradient mesh to a vector graphic having at least one node. The method may further include performing a constrained optimization of the vector graphic based on the target shape. The method may further include generating a stress metric based on a comparison of the constrained optimization and the target shape. The method may further include determining one or more unconstrained candidate vector graphics based on the stress metric. The method may further include selecting an improved vector graphic from the one or more unconstrained candidate vector graphics. The method may further include mapping the vector graphic to the improved vector graphic. The method may further include optimizing the improved vector graphic based on the target shape.
-
公开(公告)号：US20210150333A1
公开(公告)日：2021-05-20
申请号：US16686978
申请日：2019-11-18
Applicant: Adobe Inc.
Inventor： Federico Perazzi , Zhihao Xia , Michael Gharbi , Kalyan Sunkavalli
Abstract: Certain embodiments involve techniques for efficiently estimating denoising kernels for generating denoised images. For instance, a neural network receives a noisy reference image to denoise. The neural network uses a kernel dictionary of base kernels and generates a coefficient vector for each pixel in the reference image such that the coefficient vector includes a coefficient value for each base kernel in the kernel dictionary, where the base kernels are combined to generate a denoising kernel and each coefficient value indicates a contribution of a given base kernel to a denoising kernel. The neural network calculates the denoising kernel for a given pixel by applying the coefficient vector for that pixel to the kernel dictionary. The neural network applies each denoising kernel to the respective pixel to generate a denoised output image.
-
公开(公告)号：US20220156588A1
公开(公告)日：2022-05-19
申请号：US17590995
申请日：2022-02-02
Applicant: Adobe Inc.
Inventor： Federico Perazzi , Zhihao Xia , Michael Gharbi , Kalyan Sunkavalli
Abstract: Certain embodiments involve techniques for efficiently estimating denoising kernels for generating denoised images. For instance, a neural network receives a noisy reference image to denoise. The neural network uses a kernel dictionary of base kernels and generates a coefficient vector for each pixel in the reference image such that the coefficient vector includes a coefficient value for each base kernel in the kernel dictionary, where the base kernels are combined to generate a denoising kernel and each coefficient value indicates a contribution of a given base kernel to a denoising kernel. The neural network calculates the denoising kernel for a given pixel by applying the coefficient vector for that pixel to the kernel dictionary. The neural network applies each denoising kernel to the respective pixel to generate a denoised output image.
-
公开(公告)号：US11281970B2
公开(公告)日：2022-03-22
申请号：US16686978
申请日：2019-11-18
Applicant: Adobe Inc.
Inventor： Federico Perazzi , Zhihao Xia , Michael Gharbi , Kalyan Sunkavalli
Abstract: Certain embodiments involve techniques for efficiently estimating denoising kernels for generating denoised images. For instance, a neural network receives a noisy reference image to denoise. The neural network uses a kernel dictionary of base kernels and generates a coefficient vector for each pixel in the reference image such that the coefficient vector includes a coefficient value for each base kernel in the kernel dictionary, where the base kernels are combined to generate a denoising kernel and each coefficient value indicates a contribution of a given base kernel to a denoising kernel. The neural network calculates the denoising kernel for a given pixel by applying the coefficient vector for that pixel to the kernel dictionary. The neural network applies each denoising kernel to the respective pixel to generate a denoised output image.
-
公开(公告)号：US11080833B2
公开(公告)日：2021-08-03
申请号：US16692843
申请日：2019-11-22
Applicant: Adobe Inc.
Inventor： Connelly Barnes , Utkarsh Singhal , Elya Shechtman , Michael Gharbi
Abstract: A method for manipulating a target image includes generating a query of the target image and keys and values of a first reference image. The method also includes generating matching costs by comparing the query of the target image with each key of the reference image and generating a set of weights from the matching costs. Further, the method includes generating a set of weighted values by applying each weight of the set of weights to a corresponding value of the values of the reference image and generating a weighted patch by adding each weighted value of the set of weighted values together. Additionally, the method includes generating a combined weighted patch by combining the weighted patch with additional weighted patches associated with additional queries of the target image and generating a manipulated image by applying the combined weighted patch to an image processing algorithm.
-
-
-
-
-
-
-
-
-
Search Results
10
records
(took 0.024 seconds)
