Abstract: Embodiments are disclosed for machine learning-based generation of recommended layouts. The method includes receiving a set of design elements for performing generative layout recommendation. A number of each type of design element from the set of design elements is determined. A set of recommended layouts are generated using a trained generative layout model and the number and type of design elements. The set of recommended layouts are output.

Abstract: An image processing system generates an image mask from an image. The image is processed by an object detector to identify a region having an object, and the region is classified based on an object type of the object. A masking pipeline is selected from a number of masking pipelines based on the classification of the region. The region is processed using the masking pipeline to generate a region mask. An image mask for the image is generated using the region mask.

Abstract: Embodiments are disclosed for performing content linting in a graphic design system. A method of content linting includes receiving a selection of a content type to be generated, receiving a selection of a location in a digital canvas to place content of the content type, determining a placement context associated with the location in the digital canvas, identifying one or more content rules to the content based on a static analysis of the placement context, and generating, using one or more machine learning models, content of the selected content type at the location in the digital canvas using the one or more content rules.

Abstract: In accordance with the described techniques, a background generation system receives one or more images depicting an object, and textual information describing the object. A generative text model is employed to generate a prompt based on the one or more images and the textual information. Further, a generative image model is employed to generate an output image. To do so, the generative image model generates a background image based on the prompt, and the object is incorporated into the background image. Using a visual saliency model, the background generation system determines a visual saliency defining a degree of fixation on the object within the output image. The background generation system outputs the output image based on the visual saliency meeting a threshold.

Abstract: Embodiments are disclosed for recoloring a target graphic using color palettes generated using a stochastic color mapping process. One method of recoloring a target graphic using the stochastic color mapping process includes obtaining a target graphic to be recolored and a source color palette defining source colors for recoloring the target graphic. A target color set of target colors is extracted from the target graphic. The method includes computing a mapping to map source colors of a source color palette to target colors extracted from a target color set of the target graphic based on a transition probability. A destination color palette of destination colors is determined based on the mapping. The target graphic is modified by recoloring at least one object in the target graphic with a destination color from the destination color palette.

Abstract: Position-based text-to-speech model and training techniques are described. A digital document, for instance, is received by an audio synthesis service. A text-to-speech model is utilized by the audio synthesis service to generate digital audio from text included in the digital document. The text-to-speech model, for instance, is configured to generate a text encoding and a document positional encoding from an initial text sequence of the digital document. The document positional encoding is based on a location of the text encoding within the digital document. Digital audio is then generated by the text-to-speech model that includes a spectrogram having a reordered text sequence, which is different from the initial text sequence, by decoding the text encoding and the document positional encoding.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for accurately and flexibly generating modified digital images utilizing a novel swapping autoencoder that incorporates scene layout. In particular, the disclosed systems can receive a scene layout map that indicates or defines locations for displaying specific digital content within a digital image. In addition, the disclosed systems can utilize the scene layout map to guide combining portions of digital image latent code to generate a modified digital image with a particular textural appearance and a particular geometric structure defined by the scene layout map. Additionally, the disclosed systems can utilize a scene layout map that defines a portion of a digital image to modify by, for instance, adding new digital content to the digital image, and can generate a modified digital image depicting the new digital content.

Abstract: This disclosure involves using data science notebooks to customize and apply a visitor stitching framework. An event management system provides an initial visitor stitching framework via a data science notebook, wherein the data science notebook is an interactive environment for managing algorithms and data. The event management system receives, from a resource provider system via the data science notebook, a modification to the initial visitor stitching framework. The event management system applies the modification to the initial visitor stitching framework to generate a custom visitor stitching framework. The event management system processes a dataset associated with the resource provider system and a user using the custom visitor stitching framework to generate a stitched dataset associated with the user.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for performing object-specific undo and/or redo operations. For example, in one or more embodiments, the disclosed systems receive a modified digital design image comprising a first modified object and a second modified object. In some examples, the second modified object is modified after the first modified object. The disclosed systems can generate and utilize an object-specific version representation to undo an edit to the first modified object without undoing edits to the second modified object. The disclosed systems can generate and provide, for display via a user interface, an updated digital design document comprising a reverted first object and the second modified object.

Abstract: Embodiments are disclosed for generating 360-degree panoramas from input narrow field of view images. A method of generating 360-degree panoramas may include obtaining an input image and guide, generating a panoramic projection of the input image, and generating, by a panorama generator, a 360-degree panorama based on the panoramic projection and the guide, wherein the panorama generator is a guided co-modulation generator network trained to generate a 360-degree panorama from the input image based on the guide.

Abstract: Techniques are disclosed for generating image recommendations to facilitate the sale of a product. An example methodology includes identifying a product category associated with an image of the product provided by the seller, and a product sub-category associated with the product image. The method further includes retrieving one or more images of for-sale items. The retrieval is based on a search of for-sale listings using the identified product category and the identified product sub-category. The method further includes clustering the retrieved images of for-sale items into groups, each group associated with a perspective viewpoint of the for-sale item. The method further includes providing a selected image from each group as an image recommendation. The selection is based on a value score associated with each of the images of the for-sale items. A graphical status indicating completeness of the seller's image set is updated in response to recommended images being adopted.

Abstract: An object folding tool is leveraged in a digital medium environment. A two-dimensional (2D) representation of an unfolded object is obtained, and visual cues indicating folds for transforming the unfolded object into a folded object are detected. Based on the detected visual cues, a shape of the folded object is determined, and a three-dimensional (3D) representation of the folded object having the determined shape is generated. In one or more implementations, the 2D representation of the unfolded object and the 3D representation of the folded object are displayed concurrently on a display device.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for utilizing a design language model and a generative language model to generate digital design documents with design variations. In particular embodiments, the disclosed systems implement the design language model to tokenize the design of a document into a sequence of language tokens. For example, the disclosed systems tokenize visual elements and a layout of the document—in addition to optional user-added content. The generative language model utilizes the sequence of language tokens to predict a next language token representing a suggested design variation. Based on the predicted language token, the disclosed systems generate a modified digital design document visually portraying the suggested design variation. Further, in one or more embodiments, the disclosed systems perform iterative refinements to the modified digital design document.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media that generate three-dimensional hybrid mesh-volumetric representations for digital objects. For instance, in one or more embodiments, the disclosed systems generate a mesh for a digital object from a plurality of digital images that portray the digital object using a multi-view stereo model. Additionally, the disclosed systems determine a set of sample points for a thin volume around the mesh. Using a neural network, the disclosed systems further generate a three-dimensional hybrid mesh-volumetric representation for the digital object utilizing the set of sample points for the thin volume and the mesh.

Abstract: Methods, systems, and non-transitory computer readable media are disclosed for intelligently enhancing details in edited images. The disclosed system iteratively updates residual detail latent code for segments in edited images where detail has been lost through the editing process. More particularly, the disclosed system enhances an edited segment in an edited image based on details in a detailed segment of an image. Additionally, the disclosed system may utilize a detail neural network encoder to project the detailed segment and a corresponding segment of the edited image into a residual detail latent code. In some embodiments, the disclosed system generates a refined edited image based on the residual detail latent code and a latent vector of the edited image.

Abstract: 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.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for training and utilizing scale-diverse segmentation neural networks to analyze digital images at different scales and identify different target objects portrayed in the digital images. For example, in one or more embodiments, the disclosed systems analyze a digital image and corresponding user indicators (e.g., foreground indicators, background indicators, edge indicators, boundary region indicators, and/or voice indicators) at different scales utilizing a scale-diverse segmentation neural network. In particular, the disclosed systems can utilize the scale-diverse segmentation neural network to generate a plurality of semantically meaningful object segmentation outputs. Furthermore, the disclosed systems can provide the plurality of object segmentation outputs for display and selection to improve the efficiency and accuracy of identifying target objects and modifying the digital image.

Abstract: Knowledge edit techniques for text-to-image models and other generative machine learning models are described. In an example, a location is identified within a text-to-image model by a model edit system. The location is configured to influence generation of a visual attribute by a text-to-image model as part of a digital image. An edited text-to-image model is formed by editing the text-to-image model based on the location. The edit causes a change to the visual attribute in generating a subsequent digital image by the edited text-to-image model. The subsequent digital image is generated as having the change to the visual attribute by the edited text-to-image model.

Abstract: In one aspect, a processor determines a first set of video frames of a video based on a target video frame. The first set of video frames includes the target video frame, one or more frames of the video preceding the target video frame, and one or more frames of the video subsequent to the target video frame. The first set of video frames includes a sequence of video frames of the video. An encoder neural network executing on the processor encodes the first set of video frames of a video to generate a respective feature vector for each video frame in the first set. A decoder neural network executing on the processor decodes the feature vectors to generate a mask for the target video frame.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for adapting generative neural networks to target domains utilizing an image translation neural network. In particular, in one or more embodiments, the disclosed systems utilize an image translation neural network to translate target results to a source domain for input in target neural network adaptation. For instance, in some embodiments, the disclosed systems compare a translated target result with a source result from a pretrained source generative neural network to adjust parameters of a target generative neural network to produce results corresponding in features to source results and corresponding in style to the target domain.