Abstract: Methods, systems, and articles of manufacture, including computer program products, are provided for image cleanup. In some embodiments, there is provide a method which may include subsampling a first image to a first level image of a multiscale transform; performing, based on a machine learning model, an identification of a foreground portion of the first level image and a background portion of the first level image; generating, based on the identification of the foreground portion and the background portion, a first mask; upscaling the first mask to a resolution corresponding to the first image depicting the foreground item; applying the upscaled first mask to the first image to form a second image depicting the foreground item; and providing the second image depicting the foreground item to a publication system. Related systems and articles of manufacture, including computer program products, are also provided.

Abstract: In an example embodiment, a distributed storage system includes a service tier including a service node to receive a request for a logical object comprising binary data and metadata describing the binary data, and a storage tier including a plurality of storage nodes, wherein one or more of the storage nodes is to store the metadata describing the binary data. The distributed storage system also includes a coordination tier to store mapping information identifying the one or more of the storage nodes storing the metadata. The service node is also to receive the mapping information from the coordination tier, to access the metadata describing the binary data from one of the one or more of the storage nodes based on the mapping information, and to return the metadata describing the binary data in a response to the request.

Abstract: Search queries are received and search results are provided. Interaction tracking is used to determine with which search results users interact. The search results having received interactions can be represented as item vectors, which can include a vector representation of a portion of the search result, such as a title, description, or image. For each search query, the item vectors are aggregated, such as by averaging the item vectors. The search queries are stored in an item dataset as collected search queries respectively associated with the aggregate item vectors. When a new search query is received, a search query vector can be compared to the aggregate item description vectors to identify collected search queries that are related. The related collected search queries can be provided as search query recommendations or search results associated with the collected search queries can be provided in response to receiving the new search query.

Abstract: Various methods and systems for providing indications of inconsistent attributes of item listings associated in item listing videos. An item listing video—of an item listing—is accessed. The item listing video is accessed via an item listing interface of an item listing system. Extracted item features—via a machine learning engine—of an item from the item listing video, are accessed. The extracted item features are extracted based on listing-interface item features associated with listing the item. The extracted item features of the item are compared to the listing-interface item features of the item. Based on comparing the listing-interface item features to the extracted, an inconsistent attribute—between an extracted item feature and a listing-interface item feature that is associated with listing the item—is identified. An indication of an inconsistent attribute is communicated to cause display of the indication of the inconsistent attribute at the item listing interface.

Abstract: Various implementations described herein are able to leverage the interaction from one or more potential buyers relative to a digital image to automatically create a sales listing for items that appear to be of interest to the buyers. This reduces or eliminates all together the manual effort previously required of sellers in researching and collecting data on each item they wish to sell. Because of their technical nature, the innovative solutions described herein are also readily scalable which, in turn, greatly improves the seller's experience. Based on buyer interaction experiences, sales listings for each item for sale can be automatically created and listed.

Abstract: Technologies are shown for shipping route selection involving receiving sender and recipient shipping information for an item to be shipped and obtaining shipping route options for the item based on the sender and recipient shipping information. The route options are provided for display and selection of an option. A routing data block is created for the item at an address on a blockchain that stores shipping information for each stage of the selected route. A shipping tag is encoded with the blockchain address and attached to the item. The tag can be scanned to obtain the blockchain address and request information for a next stage of the shipping route from the block. The next shipping information from the block is received and utilized to ship the item to a next geolocation. The next shipping information can be determined based on current conditions, such as weather, pricing and availability.

Abstract: Colored three-dimensional digital model generation techniques and systems are described. In one example, scanning techniques are employed by a scanning system that scans a physical object while disposed within packaging to form a three-dimension digital model. A model coloring system is employed to color the three-dimensional digital model. A two-dimensional digital image is employed that captures the same or similar physical object. In one example, features of the model are matched to the image. This is then used to align a viewing perspective with respect to the model with a viewing perspective of the object within the digital image, e.g., to find which “view” of the model corresponds with the image. The color is then applied from the digital image to the model, e.g., from pixels of the image to corresponding points in the model.

Abstract: Near real-time feature simulation for online/offline point-in-time data parity is described. A computing device may assign, to respective events from a series of events, a series of time stamps associated with a near real-time (NRT) variable. The computing device may simulate a delay latency associated with processing the respective events via an online processing environment based on the series of time stamps. The computing device may provide the series of events and the simulated delay latency to a machine-learning model configured to model an outcome of the series of events using the simulated delay latency.

Abstract: Different action user-interface components in a comparison view are described. Initially, a selection is received to display a comparison view via a user interface of a listing platform. Multiple listings of the listing platform are selected for inclusion in the comparison view. A comparison view system determines which action of a plurality of actions, used by the listing platform, to associate with each of the listings. A display device displays the multiple listings concurrently in a comparison view via a user interface of the listing platform and also displays an action user-interface component (e.g., a button) in each of the plurality of listings. The action user-interface component is selectable to initiate the action associated with the respective listing. In accordance with the described techniques, the action user-interface component displayed in at least two of the multiple listings is selectable to initiate different actions in relation to the respective listing.

Abstract: Aspects of the present disclosure include a navigation system and computer-implemented methods for proactively re-routing vehicles based on an analysis of input component data obtained from the navigation-enabled devices. The navigation system scores the input component data to obtain a measure of frustration (e.g., a feeling of being upset or annoyed) of the user of the navigation-enabled device. The navigation system may provide a detour suggestion for display on the navigation-enabled device to persuade the user of the device to direct their vehicle to depart from its current location or route in an effort to remove the vehicle from traffic, and thereby reduce the frustration level of the user. The detour suggestion may include an alternative route to the original destination, or an alternative destination.

Abstract: An item sorting system is described. The item sorting system includes at least one bin that holds items to be sorted and a light source that illuminates the items stored in the at least one bin. In implementations, the at least one bin includes walls that direct light, received from the light source via a first surface, for emission via a second surface facing an interior of the at least one bin. The item sorting system further includes a recognition device that identifies the items for use in sorting the items. The item sorting system further includes an end effector that manipulates the items during sorting. In some implementations, the end effector includes tips that apply a gentle vacuum force to individually manipulate items in a manner that allows for efficient sorting without damaging the items.

Abstract: Technologies are shown for content distribution on a blockchain. An access request is received for digital content owned by a first entity and associated with one or more blocks of a blockchain. The access request is associated with a second entity. A current use condition associated with the second entity is verified to satisfy a use requirement conditions attribute defined by at least one of the one or more blocks of the blockchain. Responsive to verifying the current use condition satisfies the use requirement conditions attribute, the second entity is provided access to the digital content.

Abstract: A computing device may receive trigger data defining a trigger event and logic to add an enriched event associated with the trigger event. A computing device may detect occurrence of the trigger event in execution of one or more digital services by an executable service platform in an online operational environment of a service provider system based on the trigger data. A computing device may generate event data describing the execution of the trigger event and the enriched event responsive to the detecting based on the logic of the trigger data. A computing device may store the event data within an offline simulation environment. A computing device may simulate operation of the executable service platform offline using the event data. A computing device may output a result of the simulating for display in a user interface.

Abstract: A bad change detector for operational features of digital platforms is described. Metric data from a website is acquired by the bad change detector while implementing a first version of an operational feature of the website and while implementing a second version of the operational feature. A bad change to the website is detected by measuring an inequality among values of a frequency distribution defined by a time series of the metric data. The operational feature is reverted to the first version, automatically and without user intervention, in response to detecting the bad change, thereby improving operation of computing devices that implement the website.

Abstract: Techniques are described, as implemented by computing devices, to control access to transactions through use of cosigning based on tokenized reputation scores. This is performed by leveraging a blockchain such that tokenized reputation scores are generated based on amounts of cryptographic reputation tokens associated with blockchain account addresses associated with applicant and co-signer service provider accounts. Transactional functionality is made available to an applicant service provider account having an insufficient tokenized reputation score by using a co-signer service provider account having a sufficient tokenized reputation score to at least partially back an obligation of the applicant service provider for a transaction.

Abstract: In various example embodiments, a system and method for using camera metadata for making recommendations are presented. At least one image file having camera metadata is received. The camera metadata of the at least one image file is analyzed to determine improvements to image capture aspects associated with the at least one image file. Feedback related to the improvements to the image capture aspects associated with the at least one image file is generated. In some embodiments, the feedback may be used to generate camera and other product upgrade recommendations.

Abstract: An inventory prediction system is described that outputs a predicted inventory item not included in a user's known inventory using a cross-category directional graph that represents item categories as nodes. The inventory prediction system implements a prediction model trained using machine learning to output the predicted inventory item using the graph and at least one item from the user's known inventory. The inventory prediction system is further configured to generate a listing recommendation for the predicted inventory item. To do so, the inventory prediction system implements a logistic regression model trained using machine learning to calculate a probability that the listing recommendation should be generated using attributes of the predicted inventory item and attributes of currently trending items.

Abstract: Software is increasingly being developed as a collection of loosely coupled applications. Loosely coupled applications exchange data by publishing data to and retrieving data from a data store, such as a database, a file located on a storage cluster, etc. Data produced by one application and consumed by another is referred to as a data dependency. In some embodiments, an application's data dependencies are identified by analyzing cached query plans associated with the application. Query plans include a hierarchical representation of a query, where non-leaf nodes represent commands and leaf nodes identify data dependencies. An application's data dependencies are identified by traversing the hierarchical representation of the query. Data dependencies consumed by the application are identified by finding leaf nodes that descend from a read command, while data dependencies produced by the application are identified by finding leaf nodes that descend from a write command.

Abstract: Camera platform techniques are described. In an implementation, a plurality of digital images and data describing times, at which, the plurality of digital images are captured is received by a computing device. Objects of clothing are recognized from the digital images by the computing device using object recognition as part of machine learning. A user schedule is also received by the computing device that describes user appointments and times, at which, the appointments are scheduled. A user profile is generated by the computing device by training a model using machine learning based on the recognized objects of clothing, times at which corresponding digital images are captured, and the user schedule. From the user profile, a recommendation is generated by processing a subsequent user schedule using the model as part of machine learning by the computing device.

Abstract: Technologies are shown for trust delegation that involve receiving a first request from a subject client and responding by sending a first token having first permissions to the subject client. A second request from a first partner actor is received that includes the first token, and in response, the first partner actor is linked to the subject client in a trust stack and a second token is sent to the first actor with second permissions, the second token identifying the subject client and the first partner actor. A third request from a second partner actor is received that includes the second token, and in response, the second partner actor is linked to the first partner actor in the trust stack and a third token is sent to the second partner actor with third permissions, the third token identifying the first partner actor and the second partner actor.