Adobe, Inc.

Adobe Systems Inc.

Sohrab Amirghodsi

Elya Shechtman

Connelly Barnes

Zhe Lin

Aliakbar Darabi

Jianming Zhang

Jingwan Lu

Ning Xu

Haitian Zheng

Scott David Cohen

Sylvain Philippe Paris

Yuqian Zhou

Yilin Wang

Michal Lukac

Qing Liu

Nathan Aaron Carr

Kalyan Krishna Sunkavalli

Xin Sun

Sarah Kong

He Zhang

Su Chen

Sarah Stuckey

Yifan Jiang

Eric L Palmer

Tianshu Yu

Derek Novo

Daniel Goldman

Kevin Wampler

Zijun Wei

Florian Kainz

Ning Yu

Soheil Darabi

Yunhan Zhao

Lingzhi Zhang

Mang Tik Chiu

Xuaner Zhang

Elya Schectman

The present disclosure relates to systems, methods, and non-transitory computer readable media that generate inpainted digital images utilizing class-specific cascaded modulation inpainting neural network. For example, the disclosed systems utilize a class-specific cascaded modulation inpainting neural network that includes cascaded modulation decoder layers to generate replacement pixels portraying a particular target object class. To illustrate, in response to user selection of a replacement region and target object class, the disclosed systems utilize a class-specific cascaded modulation inpainting neural network corresponding to the target object class to generate an inpainted digital image that portrays an instance of the target object class within the replacement region. Moreover, in one or more embodiments the disclosed systems train class-specific cascaded modulation inpainting neural networks corresponding to a variety of target object classes, such as a sky object class, a water object class, a ground object class, or a human object class.

Object class inpainting in digital images utilizing class-specific inpainting neural networks

The present disclosure relates to systems, methods, and non-transitory computer readable media for generating modified digital images by utilizing a patch match algorithm to generate nearest neighbor fields for a second digital image based on a nearest neighbor field associated with a first digital image. For example, the disclosed systems can identify a nearest neighbor field associated with a first digital image of a first resolution. Based on the nearest neighbor field of the first digital image, the disclosed systems can utilize a patch match algorithm to generate a nearest neighbor field for a second digital image of a second resolution larger than the first resolution. The disclosed systems can further generate a modified digital image by filling a target region of the second digital image utilizing the generated nearest neighbor field.

Generating modified digital images utilizing nearest neighbor fields from patch matching operations of alternate digital images

The present disclosure relates to systems, methods, and non-transitory computer readable media for panoptically guiding digital image inpainting utilizing a panoptic inpainting neural network. In some embodiments, the disclosed systems utilize a panoptic inpainting neural network to generate an inpainted digital image according to panoptic segmentation map that defines pixel regions corresponding to different panoptic labels. In some cases, the disclosed systems train a neural network utilizing a semantic discriminator that facilitates generation of digital images that are realistic while also conforming to a semantic segmentation. The disclosed systems generate and provide a panoptic inpainting interface to facilitate user interaction for inpainting digital images. In certain embodiments, the disclosed systems iteratively update an inpainted digital image based on changes to a panoptic segmentation map.

Generating and providing a panoptic inpainting interface for generating and modifying inpainted digital images

Iteratively modifying inpainted digital images based on changes to panoptic segmentation maps

Learning parameters for neural networks using a semantic discriminator and an object-level discriminator

The present disclosure relates to systems, methods, and non-transitory computer readable media for accurately, efficiently, and flexibly generating harmonized digital images utilizing a self-supervised image harmonization neural network. In particular, the disclosed systems can implement, and learn parameters for, a self-supervised image harmonization neural network to extract content from one digital image (disentangled from its appearance) and appearance from another from another digital image (disentangled from its content). For example, the disclosed systems can utilize a dual data augmentation method to generate diverse triplets for parameter learning (including input digital images, reference digital images, and pseudo ground truth digital images), via cropping a digital image with perturbations using three-dimensional color lookup tables ('LUTs'). Additionally, the disclosed systems can utilize the self-supervised image harmonization neural network to generate harmonized digital images that depict content from one digital image having the appearance of another digital image.

Learning parameters for an image harmonization neural network to generate deep harmonized digital images

Embodiments are disclosed for performing wire segmentation of images using machine learning. In particular, in one or more embodiments, the disclosed systems and methods comprise receiving an input image, generating, by a first trained neural network model, a global probability map representation of the input image indicating a probability value of each pixel including a representation of wires, and identifying regions of the input image indicated as including the representation of wires. The disclosed systems and methods further comprise, for each region from the identified regions, concatenating the region and information from the global probability map to create a concatenated input, and generating, by a second trained neural network model, a local probability map representation of the region based on the concatenated input, indicating pixels of the region including representations of wires. The disclosed systems and methods further comprise aggregating local probability maps for each region.

Wire segmentation for images using machine learning

Techniques are disclosed for filling or otherwise replacing a target region of a primary image with a corresponding region of an auxiliary image. The filling or replacing can be done with an overlay (no subtractive process need be run on the primary image). Because the primary and auxiliary images may not be aligned, both geometric and photometric transformations are applied to the primary and/or auxiliary images. For instance, a geometric transformation of the auxiliary image is performed, to better align features of the auxiliary image with corresponding features of the primary image. Also, a photometric transformation of the auxiliary image is performed, to better match color of one or more pixels of the auxiliary image with color of corresponding one or more pixels of the primary image. The corresponding region of the transformed auxiliary image is then copied and overlaid on the target region of the primary image.

Image inpainting with geometric and photometric transformations

The present disclosure relates to systems, methods, and non-transitory computer readable media for training a generative inpainting neural network to accurately generate inpainted digital images via object-aware training and/or masked regularization. For example, the disclosed systems utilize an object-aware training technique to learn parameters for a generative inpainting neural network based on masking individual object instances depicted within sample digital images of a training dataset. In some embodiments, the disclosed systems also (or alternatively) utilize a masked regularization technique as part of training to prevent overfitting by penalizing a discriminator neural network utilizing a regularization term that is based on an object mask. In certain cases, the disclosed systems further generate an inpainted digital image utilizing a trained generative inpainting model with parameters learned via the object-aware training and/or the masked regularization.

Learning parameters for generative inpainting neural networks utilizing object-aware training and masked regularization

The present disclosure relates to systems, methods, and non-transitory computer readable media for accurately, efficiently, and flexibly generating modified digital images utilizing a guided inpainting approach that implements a patch match model informed by a deep visual guide. In particular, the disclosed systems can utilize a visual guide algorithm to automatically generate guidance maps to help identify replacement pixels for inpainting regions of digital images utilizing a patch match model. For example, the disclosed systems can generate guidance maps in the form of structure maps, depth maps, or segmentation maps that respectively indicate the structure, depth, or segmentation of different portions of digital images. Additionally, the disclosed systems can implement a patch match model to identify replacement pixels for filling regions of digital images according to the structure, depth, and/or segmentation of the digital images.

Generating modified digital images using deep visual guided patch match models for image inpainting

The present disclosure relates to systems, methods, and non-transitory computer readable media that generate inpainted digital images utilizing a cascaded modulation inpainting neural network. For example, the disclosed systems utilize a cascaded modulation inpainting neural network that includes cascaded modulation decoder layers. For example, in one or more decoder layers, the disclosed systems start with global code modulation that captures the global-range image structures followed by an additional modulation that refines the global predictions. Accordingly, in one or more implementations, the image inpainting system provides a mechanism to correct distorted local details. Furthermore, in one or more implementations, the image inpainting system leverages fast Fourier convolutions block within different resolution layers of the encoder architecture to expand the receptive field of the encoder and to allow the network encoder to better capture global structure.

Digital image inpainting utilizing a cascaded modulation inpainting neural network

The present disclosure relates to systems, methods, and non-transitory computer readable media that implement an inpainting framework having computer-implemented machine learning models to generate high-resolution inpainting results. For instance, in one or more embodiments, the disclosed systems generate an inpainted digital image utilizing a deep inpainting neural network from a digital image having a replacement region. The disclosed systems further generate, utilizing a visual guide algorithm, at least one deep visual guide from the inpainted digital image. Using a patch match model and the at least one deep visual guide, the disclosed systems generate a plurality of modified digital images from the digital image by replacing the region of pixels of the digital image with replacement pixels. Additionally, the disclosed systems select, utilizing an inpainting curation model, a modified digital image from the plurality of modified digital images to provide to a client device.

Generating modified digital images via image inpainting using multi-guided patch match and intelligent curation

Embodiments of the present invention provide systems, methods, and computer storage media for automatically synthesizing a content-aware sampling region for a hole-filling algorithm such as content-aware fill. Given a source image and a hole (or other target region to fill), a sampling region can be synthesized by identifying a band of pixels surrounding the hole, clustering these pixels based on one or more characteristics (e.g., color, x/y coordinates, depth, focus, etc.), passing each of the resulting clusters as foreground pixels to a segmentation algorithm, and unioning the resulting pixels to form the sampling region. The sampling region can be stored in a constraint mask and passed to a hole-filling algorithm such as content-aware fill to synthesize a fill for the hole (or other target region) from patches sampled from the synthesized sampling region.

Automatic synthesis of a content-aware sampling region for a content-aware fill

The present disclosure relates to systems, methods, and non-transitory computer-readable media for accurately restoring missing pixels within a hole region of a target image utilizing multi-image inpainting techniques based on incorporating geometric depth information. For example, in various implementations, the disclosed systems utilize a depth prediction of a source image as well as camera relative pose parameters. Additionally, in some implementations, the disclosed systems jointly optimize the depth rescaling and camera pose parameters before generating the reprojected image to further increase the accuracy of the reprojected image. Further, in various implementations, the disclosed systems utilize the reprojected image in connection with a content-aware fill model to generate a refined composite image that includes the target image having a hole, where the hole is filled in based on the reprojected image of the source image.

Image reprojection and multi-image inpainting based on geometric depth parameters

The present disclosure relates to systems, methods, and non-transitory computer readable media for accurately generating inpainted digital images utilizing a guided inpainting model guided by both plane panoptic segmentation and plane grouping. For example, the disclosed systems utilize a guided inpainting model to fill holes of missing pixels of a digital image as informed or guided by an appearance guide and a geometric guide. Specifically, the disclosed systems generate an appearance guide utilizing plane panoptic segmentation and generate a geometric guide by grouping plane panoptic segments. In some embodiments, the disclosed systems generate a modified digital image by implementing an inpainting model guided by both the appearance guide (e.g., a plane panoptic segmentation map) and the geometric guide (e.g., a plane grouping map).

Digital image inpainting utilizing plane panoptic segmentation and plane grouping

Growing community of inventors

Seattle, WA, United States of America

Sohrab Amirghodsi