Adobe, Inc.

Huawei Technologies Co., Limited

Ratheesh Kalarot

Shabnam Ghadar

Jingwan Lu

Baldo Antonio Faieta

Elya Shechtman

Wei-An Lin

Cameron Smith

Richard Zhang

Niloy Jyoti Mitra

Zhixin Shu

Jun-Yan Zhu

Akhilesh Kumar

Zhe Lin

Aliakbar Darabi

Siavash Khodadadeh

Saeid Motiian

John Thomas Nack

Sven Olsen

Nathan Aaron Carr

Oliver Wang

Holger Winnemoeller

Kevin Wampler

Christian Thomas Cantrell

Alexandru Vasile Costin

Hui Qu

Wilson Chan

Archie Samuel Bagnall

Patrick James Hebron

Jinrong Xie

Jakub Fiser

Timothy M Converse

Yannick Hold-Geoffrey

Piotr Walczyszyn

Jianming Zhang

Yannick Hold-Geoffroy

Alex Charles Filipkowski

Fatih Porikli

Tobias Hinz

Anjali Agarwal

Kevin Duarte

Alireza Zaeemzadeh

The present disclosure relates to systems, methods, and non-transitory computer readable media for generating and composting pixels of a digital image that depict hair of an individual using generative neural networks. In some embodiments, the disclosed systems receive a modification to a face crop enclosing a face depicted within a digital image. In some cases, the disclosed systems determine, from the modification, modified hair pixels within the face crop of the digital image and unmodified hair pixels outside of the face crop of the digital image. The disclosed systems generate, for the unmodified hair pixels outside of the face crop, replacement hair pixels that resemble the modified hair pixels utilizing a generative neural network. Additionally, the disclosed systems generate a modified digital image by replacing the unmodified hair pixels outside of the face crop with the replacement hair pixels.

Generating and compositing hair pixels using generative neural networks

Systems and methods for image processing are provided. Embodiments include identifying an image of a face that includes an artifact in a part of the face. A machine learning model generates an intermediate image based on the original image. The intermediate image depicts the part of the face in a closed position. Then the model generates a corrected image based on the intermediate image. The corrected image depicts the face with the part of the face in an open position and without the artifact.

Repairing irregularities in computer-generated images

Systems and methods train an encoder neural network for fast and accurate projection into the latent space of a Generative Adversarial Network (GAN). The encoder is trained by providing an input training image to the encoder and producing, by the encoder, a latent space representation of the input training image. The latent space representation is provided as input to the GAN to generate a generated training image. A latent code is sampled from a latent space associated with the GAN and the sampled latent code is provided as input to the GAN. The GAN generates a synthetic training image based on the sampled latent code. The sampled latent code is provided as input to the encoder to produce a synthetic training code. The encoder is updated by minimizing a loss between the generated training image and the input training image, and the synthetic training code and the sampled latent code.

Supervised learning techniques for encoder training

An improved system architecture uses a Generative Adversarial Network (GAN) including a specialized generator neural network to generate multiple resolution output images. The system produces a latent space representation of an input image. The system generates a first output image at a first resolution by providing the latent space representation of the input image as input to a generator neural network comprising an input layer, an output layer, and a plurality of intermediate layers and taking the first output image from an intermediate layer, of the plurality of intermediate layers of the generator neural network. The system generates a second output image at a second resolution different from the first resolution by providing the latent space representation of the input image as input to the generator neural network and taking the second output image from the output layer of the generator neural network.

Multi-scale output techniques for generative adversarial networks

Systems and methods train and apply a specialized encoder neural network for fast and accurate projection into the latent space of a Generative Adversarial Network (GAN). The specialized encoder neural network includes an input layer, a feature extraction layer, and a bottleneck layer positioned after the feature extraction layer. The projection process includes providing an input image to the encoder and producing, by the encoder, a latent space representation of the input image. Producing the latent space representation includes extracting a feature vector from the feature extraction layer, providing the feature vector lo the bottleneck layer as input, and producing the latent space representation as output. The latent space representation produced by the encoder is provided as input to the GAN, which generates an output image based upon the latent space representation.

Direct regression encoder architecture and training

The present disclosure relates to systems, methods, and non-transitory computer readable media for generating anonymized digital images utilizing a face anonymization neural network. In some embodiments, the disclosed systems utilize a face anonymization neural network to extract or encode a face anonymization guide that encodes face attribute features, such as gender, ethnicity, age, and expression. In some cases, the disclosed systems utilize the face anonymization guide to inform the face anonymization neural network in generating synthetic face pixels for anonymizing a digital image while retaining attributes, such as gender, ethnicity, age, and expression. The disclosed systems learn parameters for a face anonymization neural network for preserving face attributes, accounting for multiple faces in digital images, and generating synthetic face pixels for faces in profile poses.

Anonymizing digital images utilizing a generative adversarial neural network

An item recommendation system receives a set of recommendable items and a request to select, from the set of recommendable items, a contrast group. The item recommendation system selects a contrast group from the set of recommendable items by applying a image modification model to the set of recommendable items. The image modification model includes an item selection model configured to determine an unbiased conversion rate for each item of the set of recommendable items and select a recommended item from the set of recommendable items having a greatest unbiased conversion rate. The image modification model includes a contrast group selection model configured to select, for the recommended item, a contrast group comprising the recommended item and one or more contrast items. The item recommendation system transmits the contrast group responsive to the request.

Semantic-aware initial latent code selection for text-guided image editing and generation

The present disclosure relates to systems, non-transitory computer-readable media, and methods for combining digital images. In particular, in one or more embodiments, the disclosed systems combine latent codes of a source digital image and a target digital image utilizing a blending network to determine a combined latent encoding and generate a combined digital image from the combined latent encoding utilizing a generative neural network. In some embodiments, the disclosed systems determine an intersection face mask between the source digital image and the combined digital image utilizing a face segmentation network and combine the source digital image and the combined digital image utilizing the intersection face mask to generate a blended digital image.

Transferring faces between digital images by combining latent codes utilizing a blending network

An embodiment method includes performing first convolutional filtering on a first tensor constructed using a current frame and reference frames (or digital world reference images) of the current frame in a video, to generate a first estimated image of the current frame having a higher resolution than an image of the current frame. The method also includes performing second convolutional filtering on a second tensor constructed using the first estimated image and estimated reference images of the reference frames, to generate a second estimated image of the current having a higher resolution than the image of the current frame. The estimated reference images of the reference frames are reconstructed high resolution images of the reference images.

Multi-stage multi-reference bootstrapping for video super-resolution

The present disclosure describes systems, methods, and non-transitory computer readable media for detecting user interactions to edit a digital image from a client device and modify the digital image for the client device by using a web-based intermediary that modifies a latent vector of the digital image and an image modification neural network to generate a modified digital image from the modified latent vector. In response to user interaction to modify a digital image, for instance, the disclosed systems modify a latent vector extracted from the digital image to reflect the requested modification. The disclosed systems further use a latent vector stream renderer (as an intermediary device) to generate an image delta that indicates a difference between the digital image and the modified digital image. The disclosed systems then provide the image delta as part of a digital stream to a client device to quickly render the modified digital image.

Web-based digital image editing in real time utilizing a latent vector stream renderer and an image modification neural network

Systems and methods dynamically adjust an available range for editing an attribute in an image. An image editing system computes a metric for an attribute in an input image as a function of a latent space representation of the input image and a filtering vector for editing the input image. The image editing system compares the metric to a threshold. If the metric exceeds the threshold, then the image editing system selects a first range for editing the attribute in the input image. If the metric does not exceed the threshold, a second range is selected. The image editing system causes display of a user interface for editing the input image comprising an interface element for editing the attribute within the selected range.

Attribute control techniques for image editing

Embodiments of the present invention provide systems, methods, and computer storage media for editing images using a web-based intermediary between a user interface on a client device and an image editing neural network(s) (e.g., a generative adversarial network) on a server(s). The present image editing system supports multiple users in the same software container, advanced concurrency of projection and transformation of the same image, clubbing transformation requests from several users hosted in the same software container, and smooth display updates during a progressive projection.

Asynchronous multi-user real-time streaming of web-based image edits using generative adversarial network(s)

Systems and methods seamlessly blend edited and unedited regions of an image. A computing system crops an input image around a region to be edited. The system applies an affine transformation to rotate the cropped input image. The system provides the rotated cropped input image as input to a machine learning model to generate a latent space representation of the rotated cropped input image. The system edits the latent space representation and provides the edited latent space representation to a generator neural network to generate a generated edited image. The system applies an inverse affine transformation to rotate the generated edited image and aligns an identified segment of the rotated generated edited image with an identified corresponding segment of the input image to produce an aligned rotated generated edited image. The system blends the aligned rotated generated edited image with the input image to generate an edited output image.

Techniques for smooth region merging in image editing

Systems and methods combine an input image with an edited image generated using a generator neural network to preserve detail from the original image. A computing system provides an input image to a machine learning model to generate a latent space representation of the input image. The system provides the latent space representation to a generator neural network to generate a generated image. The system generates multiple scale representations of the input image, as well as multiple scale representations of the generated image. The system generates a first combined image based on first scale representations of the images and a first value. The system generates a second combined image based on second scale representations of the images and a second value. The system blends the first combined image with the second combined image to generate an output image.

Detail-preserving image editing techniques

Systems and methods for image processing are described. Embodiments of the present disclosure encode features of a source image to obtain a source appearance encoding that represents inherent attributes of a face in the source image; encode features of a target image to obtain a target non-appearance encoding that represents contextual attributes of the target image; combine the source appearance encoding and the target non-appearance encoding to obtain combined image features; and generate a modified target image based on the combined image features, wherein the modified target image includes the inherent attributes of the face in the source image together with the contextual attributes of the target image.

Growing community of inventors

San Jose, CA, United States of America

Ratheesh Kalarot