Abstract: Content search and geographical consideration techniques and system employed as part of a digital environment are described. In one or more implementations, a digital medium environment is described for configuring image searches by one or more computing devices. Data is received by the one or more computing devices that identifies images obtained by users and used as part of content creation, indicates geographical locations of respective said users that obtained the images or associated with the content that includes the images, and indicates times associated with the users as obtaining the images or use of the images as part of the content. A map is built by the one or more computing devices that describes how use of the images as part of the content creation is diffused over the geographical locations over the indicated times. An image search is controlled by the one or more computing devices based on the map and a geographic location associated with the image search.

Abstract: In some embodiments, a computing system computes tags for an electronic document. The computing system identifies sets of objects for the electronic document by applying a set of object-recognition rules to the electronic document, with each object-recognition rule generating a set of identified objects. The computing system generates feature maps that represent a set of identified objects. The computing system generates a heat map that identifies attributes of the electronic document including object candidates of the electronic document by applying a page-segmentation machine-learning model to the electronic document. The computing system computes a tag by applying a fusion deep learning module to the feature map and the heat map to correlate a document object identified by the feature map with an attribute of the electronic document identified by the heat map. The computing system generates the tagged electronic document by applying the tag to the electronic document.

Abstract: Stereo correspondence and depth sensor techniques are described. In one or more implementations, a depth map generated by a depth sensor is leveraged as part of processing of stereo images to assist in identifying which parts of stereo images correspond to each other. The depth map, for instance, may be utilized to assist in identifying depth discontinuities in the stereo images. Additionally, techniques may be employed to align the depth discontinuities identified from the depth map to image edges identified from the stereo images. Techniques may also be employed to suppress image edges that do not correspond to the depth discontinuities of the depth map in comparison with image edges that do correspond to the depth discontinuities as part of the identification.

Abstract: Techniques for generating a digital collage using digital images are described. An image editing system obtains a set of digital images and a collage template to be used for generating a digital collage. The image editing system calculates different arrangement permutations for placing the digital images into digital frames of the collage template, and identifies a region of interest in each of the digital images. The image editing system fits the digital images into the digital frames for each of the arrangement permutations, and calculates a permutation error value for each arrangement permutation based on proportions of the regions of interest for each of the digital images that fit into respective digital frames. The image editing system selects an arrangement permutation with an optimal permutation error value, and uses the selected arrangement permutation to arrange the set of digital images in the collage template to generate the digital collage.

Abstract: Dynamic thumbnails are described. Dynamic thumbnails provide a convenient and automated approach for providing thumbnails that are contextually relevant to a user. In at least some implementations, an input image is analyzed to generate tags describing objects or points of interest within the image, and to generate rectangles that describe the locations within the image that correspond to the generated tags. Various combinations of generated tags are analyzed to determine the smallest bounding rectangle that contains every rectangle associated with the tags in the respective combination, and a thumbnail is created. A user input is received and compared to the tags associated with the generated thumbnails, and a thumbnail that is most relevant to the user input is selected and output to the user.

Abstract: Systems and methods are disclosed for estimating aesthetic quality of digital images using deep learning. In particular, the disclosed systems and methods describe training a neural network to generate an aesthetic quality score digital images. In particular, the neural network includes a training structure that compares relative rankings of pairs of training images to accurately predict a relative ranking of a digital image. Additionally, in training the neural network, an image rating system can utilize content-aware and user-aware sampling techniques to identify pairs of training images that have similar content and/or that have been rated by the same or different users. Using content-aware and user-aware sampling techniques, the neural network can be trained to accurately predict aesthetic quality ratings that reflect subjective opinions of most users as well as provide aesthetic scores for digital images that represent the wide spectrum of aesthetic preferences of various users.

Abstract: The invention uses a PID controller engine to provide a desired distribution of targeted ad placements, including controlling campaign pace in realtime to precisely spend a campaign budget over a prescribed runtime, while producing optimum results in a cost efficient manner based on desired targeting parameters. Those parameters include one or more probabilities that a viewer associated with an ad impression opportunity: belongs to one or more targeted demographic categories; will convert with respect to a product or service being offered; has an intention to buy the product or service; or exhibits one or more defined behaviors. Probabilities are determined by a truth-based machine-learning modelling engine. Upon receiving an ad slot opportunity package, a DSP makes a bid decision and determines a bid price, and then sends a bid response to the supply-side partner, all within 200 mS of receiving the ad slot opportunity or the bid will be ignored.

Abstract: Systems and techniques are described herein for generating a triangle mesh for an image represented by curves (e.g., Bezier segments). An outline of an image is determined and reduced to a set of connected polylines that are efficiently represented in an edge list. A triangle mesh is generated based on the edge list, rather than by directly sampling the curves of the image and using the samples as vertices of triangles. Thus, the triangle mesh is generated with a number of triangles independent from a number of curves representing the image. Samples of the curves are bound to the triangle mesh by representing the samples with barycentric coordinates with respect to a triangle in the mesh. Hence, once a mesh is deformed, locations of the samples are determined from the barycentric coordinates and triangles in the deformed mesh, and used to reconstruct the curves of the deformed image.

Abstract: This disclosure relates to utilizing a statistical model trained on character dimensions to determine a likelihood of a person purchasing a product. The method may include obtaining user-input data of a first person (e.g., textual-input data, survey-response data, offer information, or clickstream data associated with a first person). A character profile for the first person is derived using the user-input data and a psycholinguistic lexicon. A statistical model is generated based on the derived character profile of the first person. Second user-input data associated with a second person is obtained. The second user-input is applied to the statistical model to determine an output of the model (e.g., a statistical probability value that quantifies, for example, a predicted intention of the second person to purchase a particular product).

Abstract: This disclosure involves generating and configuring a media player within a web browser of a digital display device. A media player is loaded within a web page executed by the digital display device. The media player initiates execution of a proxy service by registering the proxy service with the web browser. The proxy service intercepts network requests from the media player and redirects them to local storage to enable offline capability. The media player registers with a cloud media network, and in response, receives access credentials. The media player passes the access credentials to the proxy service. The media player and proxy service access the cloud media network using the access credentials to download media content and application files respectively. The proxy service uses the application files to load a new instance of the media player during a network disruption. The media player then initiates playback of the media content.

Abstract: Embodiments relate to a smart contract platform that facilitates creation, execution and verification of customized smart contracts. The smart contract platform enables design of customized smart contracts for execution and verification on a distributed ledger network, including smart contracts with logic for querying and fetching sensitive transactional data from participant nodes. A distributed ledger can store tokens indicating successful completion of one or more transaction elements without making some or all the associated transactional data visible. A smart contract form viewer can be used to view and interact with a smart contract form linked to the smart contract. The smart contract form can present contractual provisions in natural language, present transactional data to an authorized user, and accept entry or validation of designated transaction data.

Abstract: Recommendation systems and techniques are described that employ change point detection to generate recommendations for digital content. In one example, a change point detection technique is employed by a recommendation system to identify when a change point has occurred at a respective time step of a series of time steps. Detection of this change point may then be used by the recommendation system to reset the statistical model to address this change as well as generate a subsequent recommendation configured for exploration of reward distributions of the items of digital marketing content.

Abstract: An automatic teeth whitening system analyzes digital content and detects at least one teeth region in the digital content. A teeth region refers to a region or portion of the digital content that includes teeth (e.g., human teeth), and the teeth region detection includes identifying each pixel that displays part of the teeth using instance segmentation. The automatic teeth whitening system also finds the visual structure of each tooth in the teeth region using instance contours specific to the tooth. After finding the teeth region and the visual structure of each tooth in the teeth region, a whitening process is applied to the teeth to whiten them. The whitening of the teeth is performed automatically—manual steps by the user of selecting teeth regions and coloring the teeth in those regions are avoided.

Abstract: An analytics interface is provided for a content creation tool. Based on input to the analytics interface, first and second media content items are selected, where certain content is shared between the media content items. The system modifies first and second media content items to comply with formatting constraints of first and second social media services, respectively. The system associates, via a mapping, the first and second media content items with a group and publishes the first and second media content items to the first and second social media services, respectively. The system uses the mapping to match the group with first and second analytics respectively received from the first and second social media services about the first and second media content items. The system updates the analytics interface with visual representations of grouped analytics, which are generated based on matching the group with the first second analytics.

Abstract: Techniques for incorporating social-network connections information into estimated user-ratings of videos for video recommendations are described. A user-rating that a user is likely to assign to a video is predicted based on first and second predictions. The first prediction is based on ratings assigned to the video by a first set of users who have rated other videos with ratings substantially similar to the user's ratings of the other videos. The second prediction is based on ratings assigned to the video by a second set of users who have established social-network connections with the user and who have rated at least one same video that the user has previously rated. Additionally, the estimated user-rating is added to additional estimated user-ratings of other videos for comparison, and top-rated videos are identified. Then, a notification is communicated via a client device of the user to recommend the top-rated videos to the user.

Abstract: A method and apparatus for synchronizing selected attributes of a target page element on a target display layout with modified attributes of a source page element on a source display layout is disclosed. The method comprises receiving a command to synchronize a target display layout, wherein the target display layout comprises at least one target page element linked to a source page element on a source display layout; accessing a plurality of synchronization preferences, wherein the synchronization preferences define at least one selected attribute of a plurality of attributes of a target page element to be synchronized with an attribute of the linked source page element; determining whether the selected attribute of the source page element has been synchronized with the target page element; and updating the target page element when the modified attribute of the source page element has not been synchronized with the target page element.

Abstract: In some embodiments, an image manipulation application receives an incomplete image that includes a hole area lacking image content. The image manipulation application applies a contour detection operation to the incomplete image to detect an incomplete contour of a foreground object in the incomplete image. The hole area prevents the contour detection operation from detecting a completed contour of the foreground object. The image manipulation application further applies a contour completion model to the incomplete contour and the incomplete image to generate the completed contour for the foreground object. Based on the completed contour and the incomplete image, the image manipulation application generates image content for the hole area to generate a completed image.

Abstract: The present disclosure relates to a tag-based font recognition system that utilizes a multi-learning framework to develop and improve tag-based font recognition using deep learning neural networks. In particular, the tag-based font recognition system jointly trains a font tag recognition neural network with an implicit font classification attention model to generate font tag probability vectors that are enhanced by implicit font classification information. Indeed, the font recognition system weights the hidden layers of the font tag recognition neural network with implicit font information to improve the accuracy and predictability of the font tag recognition neural network, which results in improved retrieval of fonts in response to a font tag query. Accordingly, using the enhanced tag probability vectors, the tag-based font recognition system can accurately identify and recommend one or more fonts in response to a font tag query.

Abstract: In various implementations, a server is configured to execute instructions stored in storage that when executed perform operations that include receiving a hypertext transfer protocol (HTTP) request to stream a video segment of multimedia content to a client device. The video segment is of a video sub-stream of the multimedia content. The operations further include sending the video segment and an audio segment to the client device based on the HTTP request for the video segment. The sending pushes the video segment and/or the audio segment to the client device. The audio segment is of an audio sub-stream of the multimedia content. A plurality of segment sets may be pushed based on the HTTP request for the video segment. Each segment set can include an additional video segment and an additional audio segment that correspond to at least partially concurrent portions of the multimedia content.

Abstract: Techniques and systems are described that support collection of electronic signatures within an online conference. At least one user input is received as part of the online conference. The user input specifies an attendee of the online conference that is provide an electronic signature. A signature module is communicated within the online conference. The signature module is configured to receive the electronic signature from the attendee. The electronic signature received via the signature module from the attendee within the online conference is applied to the digital content.