Adobe, Inc.

Saikat Chakrabarty

Shikhar Garg

Nirmal Kumawat

Sunil Kumar

Neel Kadia

Sourabh Gupta

Jose Ignacio Echevarria Vallespi

Lars U Borg

Mrinal Kumar Sharma

Gourav Singhal

Pawan Kishor Singh

Himanshu Aggarwal

Shivani Karol

Techniques for content-aware font recommendations include obtaining an electronic document comprising an image and text. The image is processed using one or more convolutional neural networks to determine one or more image tags. The image tags are mapped to one or more font tags using a user map, a designer map, or one or more contextual synonyms of the image tags. A font to recommend for the electronic document is then determined using the one or more font tags.

Content aware font recommendation

In implementations of systems for searching for images using generated images, a computing device implements a search system to receive a natural language search query for digital images included in a digital image repository. The search system generates a set of digital images using a machine learning model based on the natural language search query. The machine learning model is trained on training data to generate digital images based on natural language inputs. The search system performs an image-based search for digital images included in the digital image repository using the set of digital images. An indication of the search result is generated for display in a user interface based on performing the image-based search.

Searching for images using generated images

A curve sampling technique for generating transformed digital visual content is leveraged in a digital medium environment. Initially, a curve sampling system obtains digital visual content, e.g., images and videos. The curve sampling system generates transformed digital visual content by transforming one or more pixels of the digital visual content using a lookup table that is derived from samples of a curve taken at evenly spaced intervals along a y-axis of a graph of the curve. Broadly speaking, the curve defines how to transform a visual characteristic of the pixels in order to achieve a desired digital visual content transformation. Additionally, the curve sampling may correspond to one step in a series of steps for transforming colors of digital visual content. Indeed, such transformations may involve multiple curve sampling steps.

Evenly spaced curve sampling technique for digital visual content transformation

The disclosure describes one or more embodiments of systems, methods, and non-transitory computer-readable media that generate a transferred hairstyle image that depicts a person from a source image having a hairstyle from a target image. For example, the disclosed systems utilize a face-generative neural network to project the source and target images into latent vectors. In addition, in some embodiments, the disclosed systems quantify (or identify) activation values that control hair features for the projected latent vectors of the target and source image. Furthermore, in some instances, the disclosed systems selectively combine (e.g., via splicing) the projected latent vectors of the target and source image to generate a hairstyle-transfer latent vector by using the quantified activation values. Then, in one or more embodiments, the disclosed systems generate a transferred hairstyle image that depicts the person from the source image having the hairstyle from the target image by synthesizing the hairstyle-transfer latent vector.

Transferring hairstyles between portrait images utilizing deep latent representations

Systems and methods for digital image processing are described. Embodiments of the systems and methods includes identifying a digital image that includes an obscured region; generating a predicted image using a neural network comprising an encoder, a feature transfer network, and a decoder, wherein the feature transfer network is trained to predict features corresponding to the obscured region based on an output of the decoder; and combining the digital image and the predicted image to produce a reconstructed version of the digital image including details corresponding to the obscured region that are not present in the digital image.

Facial reconstruction network

Embodiments are disclosed for a subtitle generator for sign language content in digital videos. In some embodiments, a method of subtitle generation includes receiving an input video comprising a representation of one or more sign language gestures, extracting landmark coordinates associated with a signer represented in the input video, determining derivative information from the landmark coordinates, and analyzing the landmark coordinates and the derivative information by at least one gesture detection model to identify a first sign language gesture.

Auto-generation of subtitles for sign language videos

Font recommendation systems and techniques are provided for determining one or more font recommendations. Content aware font recommendations can be determined for an electronic document based on information from the design context of the electronic document. For example, to generate recommended fonts for an electronic document, the font recommendation systems and techniques can obtain one or images of the electronic document and/or text from the electronic document as input, and can process the one or more images and/or the text to determine recommended font tags. In some cases, recommended fonts can be determined based on user search queries, such as when a user searches for a particular font.

A document authoring application receives a text input including one or more alphanumeric characters. A glyph extraction module in the application determines graphical data describing glyphs of the alphanumeric characters. From the graphical data, the glyph extraction module extracts visual elements, each visual element including a color or an outline from the glyphs. The glyph extraction module generates a responsive font effect that includes the extracted visual element applied to one or more of an underline effect, a strikethrough effect, a bullet point effect, or a list dot effect. Responsive to a modification of the text input, the glyph extraction module extracts an additional visual element from an additional glyph associated with the modified text. The responsive font effect is modified to include the additional visual element. The user interface is updated to display the responsive font effect or the modified font effect.

Automatic generation of responsive font effects based on font elements

Embodiments are disclosed for determining a context-aware text color recommendation for text at a text location on an image. In particular, in one or more embodiments, the disclosed systems and methods comprise obtaining an image and a text location on the image, identifying at least one color theme based on a color harmonic template associated with the image, modifying the at least one color theme based on characteristics of the image, determining accessibility for at least one color in the at least one color theme based on the text location on the image, and determining a color palette recommendation for text at the text location on the image based on the determined accessibility for the at least one color in the at least one color theme.

Context-aware text color recommendation system

In implementations of image content snapping guidelines, a guidelines segmentation system includes modules, such as an image pre-processing module to reduce the image size of a digital image if the image size exceeds an image size threshold. An object segmentation module segments objects depicted in the digital image and identifies each object by a bounding border that delineates an object region boundary. An edge detection module receives a segmented object and determines object external edges and object feature edges from the segmented object, and identifies object corners of the object. A snapping guidelines module determines image content snapping guidelines of an object depicted in the digital image, the image content snapping guidelines for an object determined based on the bounding border of the object region boundary, the object external edges, the object feature edges, and projected snapping guidelines that extend from the object corners of the object.

Image content snapping guidelines

A method, in which one or more processing devices perform operations, includes compiling a static shader at initialization of a media-processing application. The static shader is configured to transform color data from multiple pre-transform color spaces to multiple post-transform color spaces. A runtime of the media-processing application occurs subsequent to the initialization, and the method further includes determining, during the runtime, that the static shader is applicable to a color transform from a pre-transform color space to a post-transform color space. The method further includes executing the static shader to perform the color transform, based on determining that the static shader is applicable to the color transform.

Color transforms using static shaders compiled at initialization

Vector graphic font generation system implemented as part of a computing device is described. The system is configured to improve generate vector graphic font characters by detecting an object within a digital image, segmenting the digital image extract the facial region within the digital image, generating a vector graphic by converting a format of the segmented digital image into a scalable vector format, mapping the vector graphic with Unicode characters, and subsequently mapping the Unicode character with a glyph identifier. The vector graphic font generation system described herein enables the expression of a wide spectrum of emotions in numerous applications using font characters that precisely match the object, e.g., facial appearance of users as depicted in digital images.

Growing community of inventors

Noida, India

Saikat Chakrabarty