A system for monitoring surgical objects is provided. An entry scanner captures a first set of image data from a surgical object identifier. A containment surface defines a target field of view. An exit scanner captures a second set of image data from a surgical object identifier within the target field of view. A monitoring system is electrically connected to at least one of the entry and exit scanners. The monitoring system has a surgical object recognition module having a database of pre-existing surgical object identifier data. The surgical recognition module identifies at least one of the number of surgical objects, the type of surgical objects, and the number of each type of surgical objects from the first and second sets of image data by comparing the first and second sets of image data with the pre-existing surgical object identifier data.

A method of providing information about an object in a refrigerator is provided. The method includes: obtaining, using a camera, an image of a plurality of objects stored in the refrigerator; displaying the image on a display of the refrigerator; receiving a first user input indicating a first object from among the plurality of objects included in the image; and displaying a substitute image of the first object on the display based on the first user input.

A method of providing information about an object in a refrigerator is provided. The method includes: obtaining, using a camera, an image of a plurality of objects stored in the refrigerator; displaying the image on a display of the refrigerator; receiving a first user input indicating a first object from among the plurality of objects included in the image; and displaying a substitute image of the first object on the display based on the first user input.

In one implementation, a method of generating an enhanced reality (ER) map is performed by a device including one or more processors and non-transitory memory. The method includes selecting ER setting representations based on clusters of images and displaying an ER map including the ER setting representations along a path.

A process for the automatic selection of digital photographs from an archive includes the step of providing software programmed to perform at least the following steps and to perform them through such software: accessing an archive of digital photographs containing a plurality of digital photographs; setting a reference geolocation, referred to below as the “home” position; including in an identifiable selection the photographs that comply with at least the following conditions: they were taken at a spatial distance from the “home” position greater than or “greater than or equal to” a predetermined distance; and they were taken in a time sequence between each other during a timeframe where photographs taken at a distance that is less than or “less than or equal to” the predetermined distance from the “home” position are absent; preferably ordering the printing and/or archiving of the selection.

A process for the automatic selection of digital photographs from an archive includes the step of providing software programmed to perform at least the following steps and to perform them through such software: accessing an archive of digital photographs containing a plurality of digital photographs; setting a reference geolocation, referred to below as the “home” position; including in an identifiable selection the photographs that comply with at least the following conditions: they were taken at a spatial distance from the “home” position greater than or “greater than or equal to” a predetermined distance; and they were taken in a time sequence between each other during a timeframe where photographs taken at a distance that is less than or “less than or equal to” the predetermined distance from the “home” position are absent; preferably ordering the printing and/or archiving of the selection.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training an image embedding model. In one aspect, a method comprises: obtaining training data comprising a plurality of training examples, wherein each training example comprises: an image pair comprising a first image and a second image; and selection data indicating one or more of: (i) a co-click rate of the image pair, and (ii) a similar-image click rate of the image pair; and using the training data to train an image embedding model having a plurality of image embedding model parameters.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training an image embedding model. In one aspect, a method comprises: obtaining training data comprising a plurality of training examples, wherein each training example comprises: an image pair comprising a first image and a second image; and selection data indicating one or more of: (i) a co-click rate of the image pair, and (ii) a similar-image click rate of the image pair; and using the training data to train an image embedding model having a plurality of image embedding model parameters.

The Loupe system defines Loupe Visual Art DNA for art images to be presented to a user so as to maximize and customize the user experience in viewing art images delivered onto digital displays, TVs and other screens facilitating the artwork transition with and without human interaction. The Loupe system recommendations engine utilizes both human and machine curated data to determine factors of art images that will appeal to a user viewing the images. The Loupe system gathers data about visual perception, historical and academic provenance, and emotion or intention represented in an image. The gathered data is analyzed through deep learning and AI algorithms to inform recommendations and select art images to be presented to a user. The user may purchase fine art prints or select originals of the artwork image displayed, if the artist elects to make it available for sale, presented from the Loupe integrated electronic marketplace.

The Loupe system defines Loupe Visual Art DNA for art images to be presented to a user so as to maximize and customize the user experience in viewing art images delivered onto digital displays, TVs and other screens facilitating the artwork transition with and without human interaction. The Loupe system recommendations engine utilizes both human and machine curated data to determine factors of art images that will appeal to a user viewing the images. The Loupe system gathers data about visual perception, historical and academic provenance, and emotion or intention represented in an image. The gathered data is analyzed through deep learning and AI algorithms to inform recommendations and select art images to be presented to a user. The user may purchase fine art prints or select originals of the artwork image displayed, if the artist elects to make it available for sale, presented from the Loupe integrated electronic marketplace.