Abstract: A first subset of pixels of an image may be labeled with an object identifier based on user interactions with the image. Pixel data representing the pixels of the image may be passed through an embedding neural network model to generate pixel embedding vectors. A prototype embedding vector associated with the object identifier may be generated based pixel embedding vectors corresponding to the first subset of pixels. For each pixel of a second subset of pixels of the image, a probability that the pixel should be labeled with the object identifier may be determined based on the prototype embedding vector and pixel embedding vectors corresponding to the second subset of pixels. Pixels of the second subset of pixels may be labeled with the object identifier based on the determined probabilities, and the pixels in the image may be segmented based on the pixels labeled with the object identifier.

Abstract: A method includes receiving input data at a trained machine learning model that includes a common part and task-specific parts, receiving an execution instruction that identifies one or more processing tasks to be performed, processing the input data using the common part of the trained machine learning model to generate intermediate data, and processing the intermediate data using one or more of the task-specific parts of the trained machine learning model based on the execution instruction to generate one or more outputs.

Abstract: The present disclosure generally relates to methods and user interfaces for managing visual content at a computer system. In some embodiments, methods and user interfaces for managing visual content in media are described. In some embodiments, methods and user interfaces for managing visual indicators for visual content in media are described. In some embodiments, methods and user interfaces for inserting visual content in media are described. In some embodiments, methods and user interfaces for identifying visual content in media are described. In some embodiments, methods and user interfaces for translating visual content in media are described.

Abstract: Various methods for utilizing a saliency heatmaps are described. The methods include obtaining image data corresponding to an image of a scene, obtaining a saliency heatmap for the image of the scene based on a saliency network, wherein the saliency heatmap indicates a likelihood of saliency for a corresponding portion of the scene, and manipulating the image data based on the saliency heatmap. In embodiments, the saliency heatmap may be produced using a trained machine learning model. The saliency heatmap may be used for various image processing tasks, such as determining which portion(s) of a scene to base an image capture device's autofocus, auto exposure, and/or white balance operations upon. According to some embodiments, one or more bounding boxes may be generated based on the saliency heatmap, e.g., using an optimization operation, which bounding box(es) may be used to assist or enhance the performance of various image processing tasks.

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: In some implementations, a computing device may generate a day view, month views, and year view that show cards specific to each view. The cards include images, videos, and/or other assets from a media library that reflect a corresponding time frame of the card on which the assets are displayed. A selected view is presented in a graphical user interface (GUI) for interaction with a user of the media library. Upon selection of an asset displayed to the GUI, the view is switched to show more assets from a time frame similar to the selected asset while maintaining a focus on the selected asset.

Abstract: A machine may be configured to perform image evaluation of images depicting items for sale and to provide recommendations for improving the images depicting the items to increase the sales of the items depicted in the images. For example, the machine accesses a result of a user behavior analysis. The machine receives an image of an item from a user device. The machine performs an image evaluation of the received image based on an analysis of the received image and the result of the user behavior analysis. The performing of the image evaluation may include determining a likelihood of a user engaging in a desired user behavior in relation to the received image. Then, the machine generates, based on the evaluation of the received image, an output that references the received image and indicates the likelihood of a user engaging in the desired behavior.

Abstract: In some implementations, a computing device may generate a day view, month views, and year view that show cards specific to each view. The cards include images, videos, and/or other assets from a media library that reflect a corresponding time frame of the card on which the assets are displayed. A selected view is presented in a graphical user interface (GUI) for interaction with a user of the media library. Upon selection of an asset displayed to the GUI, the view is switched to show more assets from a time frame similar to the selected asset while maintaining a focus on the selected asset.

Abstract: The present disclosure generally relates to methods and user interfaces for managing visual content at a computer system. In some embodiments, methods and user interfaces for managing visual content in media are described. In some embodiments, methods and user interfaces for managing visual indicators for visual content in media are described. In some embodiments, methods and user interfaces for inserting visual content in media are described. In some embodiments, methods and user interfaces for identifying visual content in media are described. In some embodiments, methods and user interfaces for translating visual content in media are described.

Abstract: Disclosed are methods and systems for displaying items of clothing on a model having a similar body shape to that of an ecommerce user. In one aspects, a system includes one or more hardware processors configured to perform operations comprising receiving, by one or more hardware processors, an image, the image representing a user height, user weight, and user gender, causing display, by the one or more hardware processors, of a second image via a computer interface, the second image representing a model, the model selected based on a comparison of a model height, weight, and gender with the user height, weight, and gender respectively, receiving, by the one or more hardware processors, a selection of an item of clothing; and causing display, by the one or more hardware processors, of a representation of the selected model wearing the selected item of clothing.

Abstract: A machine may be configured to perform image evaluation of images depicting items for sale and to provide recommendations for improving the images depicting the items to increase the sales of the items depicted in the images. For example, the machine accesses a result of a user behavior analysis. The machine receives an image of an item from a user device. The machine performs an image evaluation of the received image based on an analysis of the received image and the result of the user behavior analysis. The performing of the image evaluation may include determining a likelihood of a user engaging in a desired user behavior in relation to the received image. Then, the machine generates, based on the evaluation of the received image, an output that references the received image and indicates the likelihood of a user engaging in the desired behavior.

Abstract: Disclosed are methods and systems for displaying items of clothing on a model having a similar body shape to that of an ecommerce user. In one aspects, a system includes one or more hardware processors configured to perform operations comprising receiving, by one or more hardware processors, an image, the image representing a user height, user weight, and user gender, causing display, by the one or more hardware processors, of a second image via a computer interface, the second image representing a model, the model selected based on a comparison of a model height, weight, and gender with the user height, weight, and gender respectively, receiving, by the one or more hardware processors, a selection of an item of clothing, and causing display, by the one or more hardware processors, of a representation of the selected model wearing the selected item of clothing.

Abstract: Various methods for utilizing a saliency heatmaps are described. The methods include obtaining image data corresponding to an image of a scene, obtaining a saliency heatmap for the image of the scene based on a saliency network, wherein the saliency heatmap indicates a likelihood of saliency for a corresponding portion of the scene, and manipulating the image data based on the saliency heatmap. In embodiments, the saliency heatmap may be produced using a trained machine learning model. The saliency heatmap may be used for various image processing tasks, such as determining which portion(s) of a scene to base an image capture device's autofocus, auto exposure, and/or white balance operations upon. According to some embodiments, one or more bounding boxes may be generated based on the saliency heatmap, e.g., using an optimization operation, which bounding box(es) may be used to assist or enhance the performance of various image processing tasks.

Abstract: Training a network for image processing with temporal consistency includes obtaining un-annotated frames from a video feed. A pretrained network is applied to the first frame of first frame set comprising a plurality of frames to obtain a first prediction, wherein the pretrained network is pretrained for a first image processing task. A current version of the pretrained network is applied to each frame of the first frame set to obtain a first prediction. A content loss term is determined, based on the first prediction and a current prediction for the frame, based on the current network. A temporal consistency loss term is also determined based on a determined consistency of pixels within each frame of the first frame set. The pretrained network may be refined based on the content loss term and the temporal term to obtain a refined network.

Abstract: Various methods for utilizing a saliency heatmaps are described. The methods include obtaining image data corresponding to an image of a scene, obtaining a saliency heatmap for the image of the scene based on a saliency network, wherein the saliency heatmap indicates a likelihood of saliency for a corresponding portion of the scene, and manipulating the image data based on the saliency heatmap. In embodiments, the saliency heatmap may be produced using a trained machine learning model. The saliency heatmap may be used for various image processing tasks, such as determining which portion(s) of a scene to base an image capture device's autofocus, auto exposure, and/or white balance operations upon. According to some embodiments, one or more bounding boxes may be generated based on the saliency heatmap, e.g., using an optimization operation, which bounding box(es) may be used to assist or enhance the performance of various image processing tasks.

Abstract: A device implementing a system for expanded search includes a processor configured to identify plural words, and generate, for each word of the plural words, a word vector based on a proximity of the word relative to other words of the plural words, the word vector comprising plural dimensions. The processor is further configured to create a compressed word vector structure comprising clusters of subsets of the plural dimensions across the word vectors, each cluster including similar values of the respective dimensions, convert the word vectors to points on at least one plane, and partition the at least one plane into nested groupings of the points based on a threshold number of points per nested grouping. The processor is further configured to create a tree look-up structure of the nested groupings, and provide the compressed word vector structure and the tree look-up structure to a client 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: Techniques are disclosed for editing captured media to overcome operational difficulties that may arise during capture operations. According to these techniques, content may be captured with a pair of cameras, a first camera having a wider field of view than a second camera. Object(s) may be detected from captured content from the wider field of view camera. The captured content may be processed from the wider field of view camera in a location of at least one detected object. Typically, operators may attempt to frame content using content from the narrower field of view camera. As a result, operators may be unaware that desired content is captured using a second, wider field of view camera. Results from the processed wider field of view data may be proposed to operators for review and, if desired, retention.

Abstract: A first subset of pixels of an image may be labeled with an object identifier based on user interactions with the image. Pixel data representing the pixels of the image may be passed through an embedding neural network model to generate pixel embedding vectors. A prototype embedding vector associated with the object identifier may be generated based pixel embedding vectors corresponding to the first subset of pixels. For each pixel of a second subset of pixels of the image, a probability that the pixel should be labeled with the object identifier may be determined based on the prototype embedding vector and pixel embedding vectors corresponding to the second subset of pixels. Pixels of the second subset of pixels may be labeled with the object identifier based on the determined probabilities, and the pixels in the image may be segmented based on the pixels labeled with the object identifier.