Abstract: The subject technology provides embodiments for performing fast corner detection in a given image for augmented reality applications. Embodiments disclose a high-speed test that examines intensities of pairs of pixels around a candidate center pixel. In one example, the examined pairs are comprised of pixels that are diametrically opposite ends of a circle formed with the candidate center pixel. Further, a pyramid of images including four rings of surrounding pixels is generated. An orientation of the pixels from the four rings are determined and a vector of discrete values of the pixels are provided. Next, a forest of trees are generated for the vector of discrete values corresponding to a descriptor for a first image. For a second image including a set of descriptors, approximate nearest neighbors are determined from the forest of tree representing closest matching descriptors from the first image.

Abstract: A system is provided for providing location-specific images to a mobile device for display. The system generally comprises three components: a mobile device having a screen, a position identification system (e.g., a GPS receiver) that determines the position of the mobile device, and a database containing location-specific images taken at various locations. Each location-specific image is associated with geographic coordinates of the location at which the image was taken. Based on the position of the mobile device as determined by the position identification system, a location-specific image is selected from the database and displayed on the screen of the mobile device. For example, when a user is using the mobile device as a car navigation system, a location-specific image of the user's destination location can be selected and displayed when the determined position of the mobile device comes within a certain distance from the geographic coordinates of the destination location.

Abstract: Various embodiments provide a method for determining color information for an image. For example, a color descriptor for an image can be determined and compared against color descriptors stored for each of a number of sample images, which each represent a color in a color space. Upon comparison, matching scores can be generated for a color match between the image and each respective sample image. In this example, the number of sample images with a matching score above a threshold value can be summed and the image can be assigned to a color associated with a highest frequency of the number of sample images. Accordingly, the assigned color of the image can then be used in a “query by color” search or a browse-by-color capability.

Abstract: Various embodiments enable an image recognition system reduce the number image match candidates before running a full-fledged pair-wise match on all image match candidates. In order to accomplish this, each inventory image can be assigned to a group. For example, a title for a book sold by an electronic marketplace could be available in multiple languages, in multiple bindings, and the book could be available in print, audio book, or electronic book. Each one of these variations could be associated with its own similarly looking inventory image, each of which could be returned as a valid match to a query image for the book. Accordingly, the inventory images for these variations could be assigned to a group for the book and, instead of geometrically processing an image for each variation, the image match system can process a single image representing all of the variations.

Abstract: Approaches introduce a pre-processing and post-processing framework to a neural network-based approach to identify items represented in an image. For example, a classifier that is trained on several categories can be provided. An image that includes a representation of an item of interest is obtained. Rotated versions of the image are generated and each of a subset of the rotated images is analyzed to determine a probability that a respective image includes an instance of a particular category. The probabilities can be used to determine a probability distribution of output category data, and the data can be analyzed to select an image of the rotated versions of the image. Thereafter, a categorization tree can then be utilized, whereby for the item of interest represented the image, the category of the item can be determined. The determined category can be provided to an item retrieval algorithm to determine primary content for the item of interest.

Abstract: A map-based interface can be used to enable a user to provide another person with funds that can be used at a particular location. In at least some embodiments, a user can select one or more buildings, locations, establishments, or other such locations and indicate an amount of money that another person is authorized to spend at that location. The funds can be provided by the user using any appropriate mechanism, such as a credit card or back account. The recipient of the funds can use information in a portable device of the user to authenticate the recipient to an establishment, which can use the dual-key information of the recipient identity and establishment location to obtain authorization to utilize the designated funds as at least partial payment for a transaction of the recipient at that establishment.

Abstract: Systems and methods track one or more points between images. A point for tracking may be selected, at least in part, on a determination of how discriminable the point is relative to other points in a region containing the point. A point of an image being tracked may be located in another image by matching a patch containing the point with another patch of the other image. A search for a matching patch may be focused in a region that is determined based at least in part on an estimate of movement of the point between images. Points may be tracked across multiple images. If an ability to track one or more points is lost, information about the points being tracked may be used to relocate the points in another image.

Abstract: Embodiments provide systems and methods for generating a street map that includes a position identifier that identifies a location on the street map. The method and system may also generate and display a plurality of images representative of the location of the position identifier. A user may interact with a position identifier or one of several scroll icons to view images of other locations on the street map and/or to obtain driving directions between two locations.

Abstract: Techniques are described for providing functionality to allow a viewer of a television show to watch a “previously on” segment of an episode of the television show and be able to watch the scenes from prior episodes referenced in the “previously on” segment.

Abstract: One or more elements on a computing device can be selected and locked from use. For example, a first user (e.g., adult) of a computing device can allow a second user (e.g., child) to use the former's device; however, the first user might not want the second user to have access to all of the elements on the device, and so the first user can select which elements he/she wants to share with the second user and which elements he/she does not want to share. For example, the first user can select elements and choose to lock the selected elements, lock all other elements, lock the selected elements for a certain period of time, or lock the selected elements but allow for earned usage, etc. The lock can be removed in response to an unlock event, which can comprise a user-initiated unlock, a timed unlock, or a user-earned locked.

Abstract: The accuracy of an image matching process can be improved by determining relevant swatch regions of the images, where those regions contain representative patterns of the items of interest represented in those images. Various processes examine a set of visual cues to determine at least one candidate object region, and then collate these regions to determine one or more representative swatch images. For apparel items, this can include locating regions such as an upper body region, torso region, clothing region, foreground region, and the like. Processes such as regression analysis or probability mapping can be used on the collated region data (along with confidence and/or probability values) to determine the appropriate swatch regions.

Abstract: Certain text editing techniques are provided to enable the user to select a few characters or words from an original collection of text by a limited number of user inputs, such as by touching or swiping a character or word to be edited. Options for probable edits of the original text can be automatically presented to the user in a manner that also only requires a minimal number of user inputs. Such techniques may facilitate text editing that is easier and more convenient for users of portable electronic devices. These techniques may be particularly advantageous for enabling editing of text acquired from an optical character recognition (OCR) process on any computing device. Other text editing shortcuts and similar approaches are also described.

Abstract: Advantages of both conventional and digital maps can be obtained by utilizing a physical map as a type of input to any of various processing devices. In various embodiments, a user can capture an image of a physical map that includes at least one location of interest. The captured image can be matched with stored map information to obtain location information corresponding to the captured image. The location information can be used to provide any of a variety of types of functionality to the user, such as providing navigation directions to that location, displaying information about attractions around that location or statistics about that location, etc. The user in some embodiments can annotate the map to provide additional input, such as areas of interest, exclusion regions, and/or proposed driving routes.

Abstract: Techniques for updating content created for a third party user are disclosed. In one embodiment, the method for updating content comprises receiving information for an item associated with a third party user, requesting a search including information for a plurality of items offered for purchase, the search request being based on the received information for the at least one item, and updating content for the third party user with one of the items included in the list of items received in response to the search request. Information for the item associated with the third party user may be received from a list of items associated with the third party user and maintained by the third party user or from a content source associated with an entity other than the third party user.

Abstract: Disclosed are various embodiments for generating recommendations utilizing visual image analysis, for example, by inserting a representative image of an item in an empty region of a digital image. Representative images for an item may be determined by generating a score for each potential image for insertion into the digital image, where a representative image is selected having a highest score from the ranking. The representative image selected may be pre-processed such that the image appears naturally in the setting.

Abstract: Various algorithms are presented that enable an image of a data matrix to be analyzed and decoded for use in obtaining information about an object or item associated with the data matrix. The algorithms can account for variations in position and/or alignment of the data matrix. In one approach, the image is analyzed to determine a connected region of pixels. The connected region of pixels can be analyzed to determine a pair of pixels, included in the connected region of pixels, that is separated a greatest distance wherein a first pixel and second pixel of the pair of pixels is associated with image coordinates. Using the image coordinates of the pair of pixels, a potential area of the image that includes the visual code can be determined and the potential area can be analyzed to verify the presence of a potential data matrix.

Abstract: Systems and methods for displaying 3D containers in a computer generated environment are described. A computing device may provide a user with a catalog of objects which may be purchased. In order to view what an object may look like prior to purchasing the object, a computing device may show a 3D container that has the same dimensions as the object. As discussed herein, the 3D container may be located and oriented based on a two-dimensional marker. Moreover, some 3D containers may contain a representation of an object, which may be a 2D image of the object.

Abstract: Various approaches provide for detecting and recognizing text to enable a user to perform various functions or tasks. For example, a user could point a camera at an object with text, in order to capture an image of that object. The camera can be integrated with a portable computing device that is capable of taking the image and processing the image (or providing the image for processing) to recognize, identify, and/or isolate the text in order to send the image of the object as well as recognized text to an application, function, or system, such as an electronic marketplace.

Abstract: Various embodiments describe systems and methods enable a computing device of a user to capture an image of a gift card, or other such monetary device containing a code, with a camera or otherwise receive an image of that gift card. The computing device can be configured to recognize codes, such as digit claim codes, of the gift card by using one or more image processing, computer vision, and/or machine learning algorithms. After a successful detection and verification of a claim code, money or funds deposited in, or otherwise available from, an account associated with the gift card can be utilized, such as applied to a purchase or deposited into the user's account. In many instances, a user interface (UI) can be provided on the computing device for the user to use to capture an image of a gift card and redeem the funds from the corresponding card.

Abstract: An object recognition system can be adapted to recognize subject matter having very few features or limited or no texture. A feature-sparse or texture-limited object can be recognized by complementing local features and/or texture features with color, region-based, shape-based, three-dimensional (3D), global, and/or composite features. Machine learning algorithms can be used to classify such objects, and image matching and verification can be adapted to the classification. Further, multiple modes of input can be integrated at various stages of the object recognition processing pipeline. These multi-modal inputs can include user feedback, additional images representing different perspectives of the object or specific regions of the object including a logo or text corresponding to the object, user behavior data, location, among others.