A computer-implemented system and method provide for a tagging user (TU) device that determines a first location of the TU device and receives, in the first location, a selection of a real-world object from a TU who views the object through the TU device. The TU device receives, from a TU, tagging information to attach to the object, and captures descriptive attributes of the object. The descriptive attributes and the tagging information associated with the first location are stored in a tagged object database.

Techniques are described herein for image recognition and retrieval. A system for image recognition and retrieval includes an imager to capture a plurality of image frames. The system also includes an image eliminator to discard any image frame of the plurality of image frames that does not meet a specified quality threshold. A remaining set of image frames is obtained. An image grouper groups the remaining set of image frames into groups of a specified number. A group of image frames is obtained. The system further includes a feature identifier to identify and describe features of the group of image frames. A requester sends a query to a content-based image retrieval server. The query contains descriptors for the features of the group of image frames.

A learning method for transforming a virtual video on a virtual world to a more real-looking video is provided. And the method includes steps of: (a) a learning device instructing a generating CNN to apply a convolutional operation to an N-th virtual training image, N-th meta data and (N-K)-th reference information to generate an N-th feature map; (b) the learning device instructing the generating CNN to apply a deconvolutional operation to the N-th feature map to generate an N-th transformed image; (c) the learning device instructing a discriminating CNN to apply a discriminating CNN operation to the N-th transformed image to generate a category score vector; (d) the learning device instructing the generating CNN to generate a generating CNN loss by referring to the category score vector and its corresponding GT, and to perform backpropagation by referring to the generating CNN loss to learn parameters of the generating CNN.

A learning method for transforming a virtual video on a virtual world to a more real-looking video is provided. And the method includes steps of: (a) a learning device instructing a generating CNN to apply a convolutional operation to an N-th virtual training image, N-th meta data and (N-K)-th reference information to generate an N-th feature map; (b) the learning device instructing the generating CNN to apply a deconvolutional operation to the N-th feature map to generate an N-th transformed image; (c) the learning device instructing a discriminating CNN to apply a discriminating CNN operation to the N-th transformed image to generate a category score vector; (d) the learning device instructing the generating CNN to generate a generating CNN loss by referring to the category score vector and its corresponding GT, and to perform backpropagation by referring to the generating CNN loss to learn parameters of the generating CNN.

Systems and methods for concept based searching or recommendation are disclosed. More particularly, embodiments of a concept based approach to the search and analysis of data, including the creation, update or use of concept networks in searching and analyzing data are disclosed, including embodiments of the usage of such concept networks in artificial intelligence systems that are capable of utilizing concepts expressed by users to return or evaluate associated images.

Techniques described herein automatically associate an image with an audio track. At least some implementations identify an audio track of interest, and automate associating an image with the audio track. Some implementations gather context information during playback of an audio track, and use the context information to automatically identify an image to associate with the audio track. Upon associating the image with the audio track, various implementations render the image during subsequent playback of the audio track.

In an embodiment, a method of emulating light sensitivity of a target includes, for each of at least some frames of a video recording, receiving an image. The method also includes accessing image metadata associated with the image. The method also includes discovering, from the image metadata, information related to an exposure setting and a gain value used by a camera in capture of the image. The method also includes deriving, from stored camera metadata for the camera, information related to a luminous flux associated with the exposure setting and the gain value used by the camera in the capture of the image. The method also includes determining an adjusted gain value corresponding to a target light sensitivity using the derived information related to the luminous flux. The method also includes generating an adjusted image using the adjusted gain value.

In an embodiment, a method of emulating light sensitivity of a target includes, for each of at least some frames of a video recording, receiving an image. The method also includes accessing image metadata associated with the image. The method also includes discovering, from the image metadata, information related to an exposure setting and a gain value used by a camera in capture of the image. The method also includes deriving, from stored camera metadata for the camera, information related to a luminous flux associated with the exposure setting and the gain value used by the camera in the capture of the image. The method also includes determining an adjusted gain value corresponding to a target light sensitivity using the derived information related to the luminous flux. The method also includes generating an adjusted image using the adjusted gain value.

A system for distinguishing a type of excreta deposited in a toilet is disclosed. The system includes a toilet and a processor. The toilet has a bowl adapted to receive multiple types of excreta from a user and a sensor which monitors the volume of excreta deposited in the toilet. The processor compares excreta volume data derived from the sensor to a database comprising excreta-type volume data and associates a time segment from the excreta volume data as representing an excreta-type. This system can provide data which may be used to determine the rate of excreta deposit into the toilet and associated those rates with excreta events types such as urination or defecation.

A system for distinguishing a type of excreta deposited in a toilet is disclosed. The system includes a toilet and a processor. The toilet has a bowl adapted to receive multiple types of excreta from a user and a sensor which monitors the volume of excreta deposited in the toilet. The processor compares excreta volume data derived from the sensor to a database comprising excreta-type volume data and associates a time segment from the excreta volume data as representing an excreta-type. This system can provide data which may be used to determine the rate of excreta deposit into the toilet and associated those rates with excreta events types such as urination or defecation.