Abstract: Various embodiments described herein utilize multiple levels of generative adversarial networks (GANs) to facilitate generation of digital images based on user-provided images. Some embodiments comprise a first generative adversarial network (GAN) and a second GAN coupled to the first GAN, where the first GAN includes an image generator and at least two discriminators, and the second GAN includes an image generator and at least one discriminator. According to some embodiments, the (first) image generator of the first GAN is trained by processing a user-provided image using the first GAN. For some embodiments, the user-provided image and the first generated image, generated by processing the user-provided image using the first GAN, are combined to produce a combined image. For some embodiments, the (second) image generator of the second GAN is trained by processing the combined image using the second GAN.

Abstract: For an input image of a person, a set of object proposals are generated in the form of bounding boxes. A pose detector identifies coordinates in the image corresponding to locations on the person's body, such as the waist, head, hands, and feet of the person. A convolutional neural network receives the portions of the input image defined by the bounding boxes and generates a feature vector for each image portion. The feature vectors are input to one or more support vector machine classifiers, which generate an output representing a probability of a match with an item. The distance between the bounding box and a joint associated with the item is used to modify the probability. The modified probabilities for the support vector machine are then compared with a threshold and each other to identify the item.

Abstract: In various example embodiments, a system and method for integration of a three-dimensional model is disclosed. In one example embodiment, a method includes receiving a plurality of images, selecting points on the images and triangulating the points to generate a plurality of depth maps, generate a three-dimensional mesh by combining the plurality of depth maps, generating a three-dimensional model of the item by projecting the plurality of images onto the mesh using the points, calibrating colors used in the model using colors diffuse properties of the colors in the images, and providing a user interface allowing a user to select one or more user points on the three-dimensional model and provide additional information associated with the selected user points.

Abstract: Example embodiments that analyze images to characterize aspects of the images rely on a same neural network to characterize multiple aspects in parallel. Because additional neural networks are not required for additional aspects, such an approach scales with increased aspects.

Abstract: Computer vision for unsuccessful queries and iterative search is described. The described system leverages visual search techniques by determining visual characteristics of objects depicted in images and describing them, e.g., using feature vectors. In some aspects, these visual characteristics are determined for search queries that are identified as not being successful. Aggregated information describing visual characteristics of images of unsuccessful search queries is used to determine common visual characteristics and objects depicted in those images. This information can be used to inform other users about unmet needs of searching users. In some aspects, these visual characteristics are used in connection with iterative image searches where users select an initial query image and then the search results are iteratively refined.

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: 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: For an input image of a person, a set of object proposals are generated in the form of bounding boxes. A pose detector identifies coordinates in the image corresponding to locations on the person's body, such as the waist, head, hands, and feet of the person. A convolutional neural network receives the portions of the input image defined by the bounding boxes and generates a feature vector for each image portion. The feature vectors are input to one or more support vector machine classifiers, which generate an output representing a probability of a match with an item. The distance between the bounding box and a joint associated with the item is used to modify the probability. The modified probabilities for the support vector machine are then compared with a threshold and each other to identify the item.

Abstract: The disclosed technologies include a robotic selling assistant that receives an item from a seller, automatically generates a posting describing the item for sale, stores the item until it is sold, and delivers or sends the item out for delivery. The item is placed in a compartment that uses one or more sensors to identify the item, retrieve supplemental information about the item, and take pictures of the item for inclusion in the posting. A seller-supplied description of the item may be verified based on the retrieved supplemental information, preventing mislabeled items from being sold.

Abstract: Systems, methods, and computer program products for identifying a candidate product in an electronic marketplace based on a visual comparison between candidate product image visual text content and input query image visual text content. Unlike conventional optical character recognition (OCR) based systems, embodiments automatically localize and isolate portions of a candidate product image and an input query image that each contain visual text content, and calculate a visual similarity measure between the respective portions. A trained neural network may be re-trained to more effectively find visual text content by using the localized and isolated visual text content portions as additional ground truths. The visual similarity measure serves as a visual search result score for the candidate product. Any number of images of any number of candidate products may be compared to an input query image to enable text-in-image based product searching without resorting to conventional OCR techniques.

Abstract: Systems, methods, and computer program products for identifying a relevant candidate product in an electronic marketplace. Embodiments perform a visual similarity comparison between candidate product image visual content and input query image visual content, process formal and informal natural language user inputs, and coordinate aggregated past user interactions with the marketplace stored in a knowledge graph. Visually similar items and their corresponding product categories, aspects, and aspect values can determine suggested candidate products without discernible delay during a multi-turn user dialog. The user can then refine the search for the most relevant items available for purchase by providing responses to machine-generated prompts that are based on the initial search results from visual, voice, and/or text inputs. An intelligent online personal assistant can thus guide a user to the most relevant candidate product more efficiently than existing search tools.

Abstract: Techniques and systems are described that leverage computer vision as part of search to expand functionality of a computing device available to a user and increase operational computational efficiency as well as efficiency in user interaction. In a first example, user interaction with items of digital content is monitored. Computer vision techniques are used to identify digital images in the digital content, objects within the digital images, and characteristics of those objects. This information is used to assign a user to a user segment of a user population which is then used to control output of subsequent digital content to the user, e.g., recommendations, digital marketing content, and so forth.

Abstract: Apparatus and method for providing contextual recommendations based on user state are disclosed herein. In some embodiments, sensor data corresponding to at least one sensor included in an item worn by a user is received. A user state is determined based on the received sensor data. In response to a state change being satisfied by at least the user state, a recommendation is determined based on the user state and a profile associated with the user. The recommendation may be presented on an electronic mobile device associated with the user.

Abstract: A machine may be configured to execute a machine-learning process for identifying and understanding fine properties of various items of various types by using images and associated corresponding annotations, such as titles, captions, tags, keywords, or other textual information applied to these images. By use of a machine-learning process, the machine may perform property identification accurately and without human intervention. These item properties may be used as annotations for other images that have similar features. Accordingly, the machine may answer user-submitted questions, such as “What do rustic items look like?,” and items or images depicting items that are deemed to be rustic can be readily identified, classified, ranked, or any suitable combination thereof.

Abstract: A machine is configured to determine fashion preferences of users and to provide item recommendations based on the fashion preferences. For example, the machine accesses an indication of a fashion style of a user. The fashion style is determined based on automatically captured data pertaining to the user. The machine identifies, based on the fashion style, one or more fashion items from an inventory of fashion items. The machine generates one or more selectable user interface elements for inclusion in a user interface. The one or more user interface elements correspond to the one or more fashion items. The machine causes generation and display of the user interface that includes the one or more selectable user interface elements. A selection of a selectable user interface element results in display of a combination of an image of a particular fashion item and an image of an item worn by the user.

Abstract: Various embodiments use a neural network to analyze images for aspects that characterize the images, to present locations of those aspects on the images, and, additionally, to permit a user to interact with those locations on the images. For example, a user may interact with a visual cue over one of those locations to modify, refine, or filter the results of a visual search, performed on a publication corpus, that uses an input image (e.g., one captured using a mobile device) as a search query.

Abstract: Methods, systems, and computer programs are presented for adding new features to a network service. A method includes receiving an image depicting an object of interest. A category set is determined for the object of interest and an image signature is generated for the image. Using the category set and the image signature, the method identifies a set of publications within a publication database and assigns a rank to each publication. The method causes presentation of the ranked list of publications at a computing device from which the image was received.

Abstract: The disclosed technologies include a robotic selling assistant that receives an item from a seller, automatically generates a posting describing the item for sale, stores the item until it is sold, and delivers or sends the item out for delivery. The item is placed in a compartment that uses one or more sensors to identify the item, retrieve supplemental information about the item, and take pictures of the item for inclusion in the posting. A seller-supplied description of the item may be verified based on the retrieved supplemental information, preventing mislabeled items from being sold.

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: Electronic content that has a tactile dimension when presented on a tactile-enabled computing device may be referred to as tactile-enabled content. A tactile-enabled device is a device that is capable of presenting tactile-enabled content in a manner that permits a user to experience tactile quality of electronic content. In one example embodiment, a system is provided for generating content that has a tactile dimension when presented on a tactile-enabled device.