Abstract: In some embodiments, a review application accesses a review file containing a text-based comment describing a change to an object in a design file. The design file is created and editable in a design application different from the review application. The review application determines, from the text-based comment, a tool in the design application configured to implement the change to the object described in the text-based comment. The review application updates the user interface to include a user interface of the tool along with the text-based comment and receives, through the user interface of the tool, a confirmation of applying the change to the object in the design file. The review application further causes the change to the object to be applied to the design file.

Abstract: Methods, apparatus, and systems are provided for obfuscating facial identity in images by synthesizing a new facial image for an input image. A base face is detected from or selected for an input image. Facial images that are similar to the base face are selected and combined to create a new facial image. The new facial image is added to the input image such that the input image includes a combination of the base face and the new facial image. Where no base face is detected in the input image, a base face is selected from reference facial images based at least on pose keypoints identified in the input image. After a new facial image is generated based on the selected base face, a combination of the new facial image and the base facial image are added to the input image by aligning one or more pose keypoints.

Abstract: Digital content text processing techniques are described. In one example, a text corpus is extracted from digital content and text corpus keywords are identified that are included in the text corpus. A plurality of clusters is formed based on the text corpus keywords. Cluster scores are generated for reviews that define a probability the review belongs to a respective cluster, e.g., based on review keywords extracted from the reviews. Sentiment values and sentiment weights are also generated. The sentiment values describe a sentiment that each of the reviews exhibits towards a respective cluster, e.g., a type of sentiment such as positive, neutral, or negative. The sentiment weights describe an amount of weight to be applied for each sentiment with respect to that cluster. The service provider system then generates ranking scores based on the cluster scores and the sentiment scores which are used to control output of the reviews.

Abstract: Methods, systems, and non-transitory computer readable media are disclosed for generating a composite image comprising objects in positions from two or more different digital images. In one or more embodiments, the disclosed system receives a sequence of images and identifies objects within the sequence of images. In one example, the disclosed system determines a target position for a first object based on detecting user selection of the first object in the target position from a first image. The disclosed system can generate a fixed object image comprising the first object in the target position. The disclosed system can generate preview images comprising the fixed object image with the second object sequencing through a plurality of positions as seen in the sequence of images. Based on a second user selection of a desired preview image, the disclosed system can generate the composite image.

Abstract: A framework for identifying prominent influencers/celebrities on social media webpages and customizing selection of thumbnails for presentation on social media webpages is provided. A most prominent influencer/celebrity associated with a social media webpage is determined. A presented thumbnail associated with each of a plurality of videos on a social media webpage is identified, each thumbnail including a picture/image of the most prominent influencer/celebrity. A feature vector is created for each identified presented thumbnail. Also identified is a plurality of potential thumbnails, each representative of a common video to be posted on the webpage and including an image of the identified most prominent influencer/celebrity. The feature vectors for the identified presented thumbnails and potential thumbnails are grouped into vector clusters. A centroid vector is identified for each vector cluster and a most prominent cluster is identified.

Abstract: An image differentiation system receives input feature vectors for multiple input images and reference feature vectors for multiple reference images. In some cases, the feature vectors are extracted by an image feature extraction module trained based on training image triplets. A differentiability scoring module determines a differentiability score for each input image based on a distance between the input feature vectors and the reference feature vectors. The distance for each reference feature vector is modified by a weighting factor based on interaction metrics associated with the corresponding reference image. In some cases, an input image is identified as a differentiated image based on the corresponding differentiability score. Additionally or alternatively, an image modification module determines an image modification that increases the differentiability score of the input image. The image modification module generates a recommended image by applying the image modification to the input image.

Abstract: Techniques are disclosed for identifying asides within a document, and detecting a display order of contents based of the identified asides. In a document, an “aside” represents a content region of the document that is distinct from the main content regions, and may be visually distinguishable from the main content region. In an example, a document is received, where the document lacks identification of asides. The document is analyzed to identify asides within the document. A display order of contents within the document is then determined, based on the identified asides. For example, in the display order, the asides are ordered between two segments of the main content and/or at a beginning or an end of the main content, but may not be ordered to be embedded in between a segment of the main content. The document is displayed in accordance with the display order.

Abstract: A method of generating accessible content is described. Embodiments of the method identifies a plurality of channels for a multimedia communication session, generate a master timeline for the communication session, wherein the master timeline comprises a chronological ordering of events from each of the channels, and wherein each of the events is associated with event-specific audio data, and present the multimedia communication session to a user to enable the user to transition among the channels based on the master timeline.

Abstract: In some embodiments, a review application accesses a review file containing a text-based comment describing a change to an object in a design file. The design file is created and editable in a design application different from the review application. The review application determines, from the text-based comment, a tool in the design application configured to implement the change to the object described in the text-based comment. The review application updates the user interface to include a user interface of the tool along with the text-based comment and receives, through the user interface of the tool, a confirmation of applying the change to the object in the design file. The review application further causes the change to the object to be applied to the design file.

Abstract: In implementations of a visitor engagement detection and control system for online sessions with a listing platform, a computing device implements a visitor engagement system to monitor user actions by the visitors with items listed by the listing platform, and map the user actions and the items listed by the listing platform to a timeline of the online session. An engagement module determines levels of visitor engagement in the online session based on the mapping of the user actions and the items, and initiates communication with a low-engagement visitor during the online session to increase an engagement level of the low-engagement visitor. In some cases, a communication is then established between the low-engagement visitor and a different low-engagement visitor, an engaged visitor, or an administrator of the listing platform.

Abstract: In implementations of jointly editing related objects in a digital image, an image editing application segments a first object in a digital image, and an additional object corresponding to the first object, such a shadow cast by the first object, a reflection of the first object, or an object of a same object class as the first object. Respective stick diagrams for the first object and the additional object are generated, and a mapping of the first object to the additional object is generated based on the stick diagrams. Based on a user request to edit the first object, such as to warp the first object, the first object and the additional object are jointly edited based on the mapping. Accordingly, realistic digital images are efficiently generated that maintain correspondence between related objects, without requiring a user to edit each object individually, thus saving time and resources.

Abstract: Techniques are provided for customizing, based on a user's activity over time, the selection of a video thumbnail for inclusion as a selectable interface element or element of a graphical interface. A server computer identifies events associated with prior interactions of a user and computes a time-decayed metric based on the time of a predicted future action of the user in comparison to a respective time of each identified event. Based on the time-decayed metric, the server computer selects a video thumbnail that is more relevant to the first event than the second event for presentation to the user.

Abstract: In implementations of systems for role classification, a computing device implements a role system to receive data describing a corpus of text that is associated with a user ID. Feature values of features are generated by a first machine learning model by processing the corpus of text, the features representing questions with respect to the corpus of text and the feature values representing answers to the questions included in the corpus of text. A classification of a role is generated by a second machine learning model by processing the feature values, the classification of the role indicating a relationship of the user ID with respect to a product or service. The role system outputs an indication of the classification of the role for display in a user interface of a display device.

Abstract: Certain embodiments involve visually augmenting images of three-dimensional containers with virtual elements that fill one or more empty regions of the three-dimensional containers. For instance, a computing system receives a first image that depicts a storage container and identify sub-containers within the storage container. The computing system selects, from a virtual object library, a plurality of virtual objects that are semantically related to the sub-container. The computing system determines an arrangement of the virtual objects within the sub-container based on semantics associated with the sub-container and the plurality of virtual objects. The computing system generates a second image that depicts the arrangement of the plurality of virtual objects within the storage container and sub-containers. The computing system generates, for display, the second image depicting the storage container and the arrangement of the virtual objects.

Abstract: Digital image selection techniques are described that employ machine learning to select a digital image of an object from a plurality of digital images of the object. The plurality of digital images each capture the object for inclusion as part of generating digital content, e.g., a webpage, a thumbnail to represent a digital video, and so on. In one example, digital image selection techniques are described that employ machine learning to select a digital image of an object from a plurality of digital images of the object. As a result, the service provider system may select a digital image of an object from a plurality of digital images of the object that has an increased likelihood of achieving a desired outcome and may address the multitude of different ways in which an object may be presented to a user.

Abstract: The present disclosure relates to a computer-implemented method for generating an enhanced image from an original image, the method including segmenting the original image into a segmented image using an artificial neural network; curve fitting the segmented image to determine boundary artifacts; removing the determined boundary artifacts to generate a smoothed boundary image; and generating the enhanced image from the original image and the smoothed boundary image. The image maybe enhanced further by correcting for glare and adding artificial light.

Abstract: Techniques are disclosed for narrowing search requests, based on interaction between a search system and a user. For example, a plurality of search results is generated in response to a search query. To reduce the number of search results, a plurality of attributes or features of the search results are identified. Each feature has a corresponding plurality of clusters, where a cluster of a feature represents a corresponding range or value of the feature. For each feature, the first plurality of search results is categorized into the corresponding plurality of clusters of the corresponding feature. A feature is then selected. The search system interacts with the user, to identify a cluster of the plurality of clusters of the selected feature in which one or more intended search results belong. Based on such identification of the cluster, the search system refines or narrows down the first plurality of search results.

Abstract: Certain embodiments involve visually augmenting images of three-dimensional containers with virtual elements that fill one or more empty regions of the three-dimensional containers. For instance, a computing system receives a first image that depicts a storage container and identify sub-containers within the storage container. The computing system selects, from a virtual object library, a plurality of virtual objects that are semantically related to the sub-container. The computing system determines an arrangement of the virtual objects within the sub-container based on semantics associated with the sub-container and the plurality of virtual objects. The computing system generates a second image that depicts the arrangement of the plurality of virtual objects within the storage container and sub-containers. The computing system generates, for display, the second image depicting the storage container and the arrangement of the virtual objects.

Abstract: Techniques for propagating a reflection of an object. In an example, a method includes receiving an input image comprising a first reflection of a first object on a reflective surface. The method further includes generating a second reflection for a second object in the input image. The second reflection is a reflection of the second object on the reflective surface. The method includes adding the second reflection to the input image. The method includes outputting a modified image comprising the first object, first reflection, the second object, and the second reflection.

Abstract: In implementations of user interaction determination within a webinar system, a computing device implements a webinar system that exposes interactive elements on user devices during a webinar and monitors device interactions reflecting user interactions with webinar content on the user devices. The webinar system determines amounts of user interaction based on the device interactions, and can output the interactive elements based on the device interactions. The webinar system can receive user responses to the interactive elements, and maintain a pipeline that assigns levels to the users based on the user responses and the interactive elements. Users are determined as sales leads based on the levels for the users in the pipeline.