Adobe, Inc.

Zhifei Zhang

Zhe Lin

Zhaowen Wang

Scott David Cohen

Hailin Jin

Matthew David Fisher

Zhihong Ding

Jason Wen Yong Kuen

Sijie Zhu

Jianming Zhang

Brian Lynn Price

Jonathan Worthen Brandt

Kevin Gary Smith

Jingwan Lu

Jose Ignacio Echevarria Vallespi

Richard Zhang

Yijun Li

Nirmal Kumawat

Niloy Jyoti Mitra

Krishna Kumar Singh

Stephen Diverdi

Qing Liu

Yulun Zhang

Darshan D Prasad

Xuan Li

Xingqian Xu

Chinthala Pradyumna Reddy

Sangryul Jeon

Daniil Pakhomov

Mariette Souppe

Ohiremen Dibua

Tobias Hinz

Shaoan Xie

Systems and methods for multi-modal image editing are provided. In one aspect, a system and method for multi-modal image editing includes identifying an image, a prompt identifying an element to be added to the image, and a mask indicating a first region of the image for depicting the element. The system then generates a partially noisy image map that includes noise in the first region and image features from the image in a second region outside the first region. A diffusion model generates a composite image map based on the partially noisy image map and the prompt. In some cases, the composite image map includes the target element in the first region that corresponds to the mask.

Multi-modal image editing

The present disclosure relates to systems, methods, and non-transitory computer readable media that utilizes artificial intelligence to learn to recommend foreground object images for use in generating composite images based on geometry and/or lighting features. For instance, in one or more embodiments, the disclosed systems transform a foreground object image corresponding to a background image using at least one of a geometry transformation or a lighting transformation. The disclosed systems further generating predicted embeddings for the background image, the foreground object image, and the transformed foreground object image within a geometry-lighting-sensitive embedding space utilizing a geometry-lighting-aware neural network. Using a loss determined from the predicted embeddings, the disclosed systems update parameters of the geometry-lighting-aware neural network. The disclosed systems further provide a variety of efficient user interfaces for generating composite digital images.

Recommending objects for image composition using geometry-and-lighting aware search and efficient user interface workflows

Recommending objects for image composition using a geometry-and-lighting aware neural network

The present disclosure relates to systems, methods, and non-transitory computer-readable media that modify digital images via multi-layered scene completion techniques facilitated by artificial intelligence. For instance, in some embodiments, the disclosed systems receive a digital image portraying a first object and a second object against a background, where the first object occludes a portion of the second object. Additionally, the disclosed systems pre-process the digital image to generate a first content fill for the portion of the second object occluded by the first object and a second content fill for a portion of the background occluded by the second object. After pre-processing, the disclosed systems detect one or more user interactions to move or delete the first object from the digital image. The disclosed systems further modify the digital image by moving or deleting the first object and exposing the first content fill for the portion of the second object.

Modifying digital images via multi-layered scene completion facilitated by artificial intelligence

The present disclosure relates to systems, methods, and non-transitory computer-readable media that implements related image search and image modification processes using various search engines and a consolidated graphical user interface. For instance, in one or more embodiments, the disclosed systems receive an input digital image and search input and further modify the input digital image using the image search results retrieved in response to the search input. In some cases, the search input includes a multi-modal search input having multiple queries (e.g., an image query and a text query), and the disclosed systems retrieve the image search results utilizing a weighted combination of the queries. In some implementations, the disclosed systems generate an input embedding for the search input (e.g., the multi-modal search input) and retrieve the image search results using the input embedding.

Generating unified embeddings from multi-modal canvas inputs for image retrieval

Systems, methods, and non-transitory computer-readable media implements related image search and image modification processes using various search engines and a consolidated graphical user interface. For instance, one or more embodiments involve receiving an input digital image and search input and further modifying the input digital image using the image search results retrieved in response to the search input. In some cases, the search input includes a multi-modal search input having multiple queries (e.g., an image query and a text query), and one or more embodiments involve retrieving the image search results utilizing a weighted combination of the queries. Some implementations involve generating an input embedding for the search input (e.g., the multi-modal search input) and retrieving the image search results using the input embedding.

Generating embeddings for text and image queries within a common embedding space for visual-text image searches

Systems and methods for image processing are configured. Embodiments of the present disclosure encode a content image and a style image using a machine learning model to obtain content features and style features, wherein the content image includes a first object having a first appearance attribute and the style image includes a second object having a second appearance attribute; align the content features and the style features to obtain a sparse correspondence map that indicates a correspondence between a sparse set of pixels of the content image and corresponding pixels of the style image; and generate a hybrid image based on the sparse correspondence map, wherein the hybrid image depicts the first object having the second appearance attribute.

Exemplar-based object appearance transfer driven by correspondence

The present disclosure relates to systems, non-transitory computer-readable media, and methods for transferring global style features between digital images utilizing one or more machine learning models or neural networks. In particular, in one or more embodiments, the disclosed systems receive a request to transfer a global style from a source digital image to a target digital image, identify at least one target object within the target digital image, and transfer the global style from the source digital image to the target digital image while maintaining an object style of the at least one target object.

Applying object-aware style transfer to digital images

The present disclosure relates to systems, methods, and non-transitory computer readable media for accurately and flexibly generating scalable and semantically editable font representations utilizing a machine learning approach. For example, the disclosed systems generate a font representation code from a glyph utilizing a particular neural network architecture. For example, the disclosed systems utilize a glyph appearance propagation model and perform an iterative process to generate a font representation code from an initial glyph. Additionally, using a glyph appearance propagation model, the disclosed systems automatically propagate the appearance of the initial glyph from the font representation code to generate additional glyphs corresponding to respective glyph labels. In some embodiments, the disclosed systems propagate edits or other changes in appearance of a glyph to other glyphs within a glyph set (e.g., to match the appearance of the edited glyph).

Generating scalable and semantically editable font representations

The present disclosure relates to systems, methods, and non-transitory computer-readable media for accurately and flexibly generating scalable fonts utilizing multi-implicit neural font representations. For instance, the disclosed systems combine deep learning with differentiable rasterization to generate a multi-implicit neural font representation of a glyph. For example, the disclosed systems utilize an implicit differentiable font neural network to determine a font style code for an input glyph as well as distance values for locations of the glyph to be rendered based on a glyph label and the font style code. Further, the disclosed systems rasterize the distance values utilizing a differentiable rasterization model and combines the rasterized distance values to generate a permutation-invariant version of the glyph corresponding glyph set.

Generating scalable fonts utilizing multi-implicit neural font representations

This disclosure describes one or more implementations of systems, non-transitory computer-readable media, and methods that extract a texture from embedded text within a digital image utilizing kerning-adjusted glyphs. For example, the disclosed systems utilize text recognition and text segmentation to identify and segment glyphs from embedded text depicted in a digital image. Subsequently, in some implementations, the disclosed systems determine optimistic kerning values between consecutive glyphs and utilize the kerning values to reduce gaps between the consecutive glyphs. Furthermore, in one or more implementations, the disclosed systems generate a synthesized texture utilizing the kerning-value-adjusted glyphs by utilizing image inpainting on the textures corresponding to the kerning-value-adjusted glyphs. Moreover, in certain instances, the disclosed systems apply a target texture to a target digital text based on the generated synthesized texture.

Extracting textures from text based images

Systems and methods for text segmentation are described. Embodiments of the inventive concept are configured to receive an image including a foreground text portion and a background portion, classify each pixel of the image as foreground text or background using a neural network that refines a segmentation prediction using a key vector representing features of the foreground text portion, wherein the key vector is based on the segmentation prediction, and identify the foreground text portion based on the classification.

Text refinement network

The present disclosure relates to systems, computer-implemented methods, and non-transitory computer readable medium for automatically transferring makeup from a reference face image to a target face image using a neural network trained using semi-supervised learning. For example, the disclosed systems can receive, at a neural network, a target face image and a reference face image, where the target face image is selected by a user via a graphical user interface (GUI) and the reference face image has makeup. The systems transfer, by the neural network, the makeup from the reference face image to the target face image, where the neural network is trained to transfer the makeup from the reference face image to the target face image using semi-supervised learning. The systems output for display the makeup on the target face image.

Automatic makeup transfer using semi-supervised learning

Systems and methods provide for generating glyph initiations using a generative font system. A glyph variant may be generated based on an input vector glyph. A plurality of line segments may be approximated using a differentiable rasterizer with the plurality of line segments representing the contours of the glyph variant. A bitmap of the glyph variant may then be generated based on the line segments. The image loss between the bitmap and a rasterized representation of a vector glyph may be calculated and provided to the generative font system. Based on the image loss, a refined glyph variant may be provided to a user.

Differentiable rasterizer for vector font generation and editing

Systems and methods for texture hallucination with a large upscaling factor are described. Embodiments of the systems and methods may receive an input image and a reference image, extract an upscaled feature map from the input image, match the input image to a portion of the reference image, wherein a resolution of the reference image is higher than a resolution of the input image, concatenate the upscaled feature map with a reference feature map corresponding to the portion of the reference image to produce a concatenated feature map, and generate a reconstructed image based on the concatenated feature map using a machine learning model trained with a texture loss and a degradation loss, wherein the texture loss is based on a high frequency band filter, and the degradation loss is based on a downscaled version of the reconstructed image.

Growing community of inventors

San Jose, CA, United States of America

Zhifei Zhang