Abstract: The simplification of neural network models is described. For example, a method for simplifying a neural network model includes providing the neural network model to be simplified, defining a first temporal filter for the conveyance of input from a neuron to an other spatially-extended neuron along the arborized projection, defining a second temporal filter for the conveyance of input from yet another neuron to the spatially-extended neuron along the arborized projection, replacing, in the neural network model, the first, spatially-extended neuron with a first, spatially-constrained neuron and the arborized projection with a first connection extending between the first, spatially-constrained neuron and the second neuron, wherein the first connection filters input from the second neuron in accordance with the first temporal filter and a second connection extending between the first spatially-constrained neuron and the third neuron.

Abstract: A computing device is configured to allocate memory for exclusive use of an execution entity from both a shared memory area and a private memory area of the device. Specifically, the shared memory area is configured with a united memory pool (UMP) component. The UMP component is configured to provide portions of huge page-based memory to execution entities for exclusive use of the execution entities. Memory granules that are allocated to the UMP component are divided into smaller memory chunks (which are smaller than a huge page), each of which can be allocated for exclusive use of an execution entity. These memory chunks are mapped to virtual address spaces of the assigned execution entities. Because memory granules can be allocated to, and deallocated from, the UMP component at run-time, the amount of memory that is available for private data generated by execution entities is able to be dynamically adjusted.

Abstract: A pallet system for the stacking and movement of a product, such as ammunition cans, is described. The pallet system includes vertical supports that substantially contribute to the strength and integrity of the pallet system. The pallet system allows for relatively rapid deployment of the pallet system's contents.

Abstract: Techniques are shown for generating image frames from a media presentation. In one embodiment a computer implemented method is provided. The method includes identifying, by a processing device, image frames from a media presentation comprising a plurality of image frames. Candidate thumbnails are selected from the identified image frames. A probability is determined that a selected candidate thumbnail with a success ranking higher than other selected thumbnails is an optimum candidate thumbnail for representing the media presentation in view of a relationship between the success ranking of the selected candidate thumbnail and the success rankings of the other selected candidate thumbnails. Thereupon, data for displaying the selected candidate thumbnail to a user as a representative of the media presentation is provided by the processing device.

Abstract: Techniques are described herein for effectively managing usage of a shared object cache in a container database management system (DBMS). The shared object cache maintains shared objects belonging to a set of pluggable databases (PDBs) hosted by the container DBMS. In an embodiment, a shared object metadata extension structure (SOMEX) is maintained for each PDB. The SOMEX stores metadata for each shared object of the PDB that resides in the shared object cache. In an embodiment, a share of the shared object cache is maintained for shared objects from each PDB in the set of PDBs based on entries in the SOMEX for the PDB.

Abstract: The subject disclosure relates to a machine-implemented method handling context data for tagged messages. The method includes identifying a tagged message, wherein the tagged message is generated by a user and includes content, obtaining context data associated with the tagged message, wherein the context data is obtained based on at least one of user input for the tagged message or server data for the tagged message, and providing access to the tagged message according to the obtained context data.

Abstract: Techniques are provided herein for generating an integral image of an input image in parallel across the cores of a multi-core processor. The input image is split into a plurality of tiles, each of which is stored in a scratchpad memory associated with a distinct core. At each tile, a partial integral image of the tile is first computed over the tile, using a Single-Pass Algorithm. This is followed by aggregating partial sums belonging to subsets of tiles using a 2D Inclusive Parallel Prefix Algorithm. A summation is finally performed over the aggregated partial sums to generate the integral image over the entire input image.

Abstract: A measuring vessel (1) for determining the accuracy with which a liquid fuel dispenser dispenses liquid fuel comprises a neck (4) of transparent material. A primary arcuate panel (16) having three primary graduated scales (14) corresponding to three different kinds of liquid fuels is secured to the neck (4). Primary graduations (17) of the respective primary graduated scales (14) indicate the level to which the meniscus of actual volumes of the respective liquid fuel dispensed at the temperatures corresponding to the primary graduations (17) should rise so that the energy content of the actual volumes of the respective liquid fuels dispensed is equal to the energy content of a predefined calibration volume of the corresponding liquid fuel at a predefined reference temperature typically of 15° C.

Abstract: Techniques are described herein for effectively managing usage of a shared object cache in a container database management system (DBMS). The shared object cache maintains shared objects belonging to a set of pluggable databases (PDBs) hosted by the container DBMS. In an embodiment, a shared object metadata extension structure (SOMEX) is maintained for each PDB. The SOMEX stores metadata for each shared object of the PDB that resides in the shared object cache. In an embodiment, a share of the shared object cache is maintained for shared objects from each PDB in the set of PDBs based on entries in the SOMEX for the PDB.

Abstract: A measuring vessel (1) for determining the accuracy with which a liquid fuel dispenser dispenses liquid fuel comprises a neck (4) of transparent material. A primary arcuate panel (16) having three primary graduated scales (14) corresponding to three different kinds of liquid fuels is secured to the neck (4). Primary graduations (17) of the respective primary graduated scales (14) indicate the level to which the meniscus of actual volumes of the respective liquid fuel dispensed at the temperatures corresponding to the primary graduations (17) should rise so that the energy content of the actual volumes of the respective liquid fuels dispensed is equal to the energy content of a predefined calibration volume of the corresponding liquid fuel at a predefined reference temperature typically of 15° C.

Abstract: Techniques are shown for generating image frames from a media presentation. In one embodiment a computer implemented method is provided. The method includes identifying, by a processing device, image frames from a media presentation comprising a plurality of image frames. Candidate thumbnails are selected from the identified image frames. A probability is determined that a selected candidate thumbnail with a success ranking higher than other selected thumbnails is an optimum candidate thumbnail for representing the media presentation in view of a relationship between the success ranking of the selected candidate thumbnail and the success rankings of the other selected candidate thumbnails. Thereupon, data for displaying the selected candidate thumbnail to a user as a representative of the media presentation is provided by the processing device.

Abstract: The subject disclosure relates to a machine-implemented method handling context data for tagged messages. The method includes identifying a tagged message, wherein the tagged message is generated by a user and includes content, obtaining context data associated with the tagged message, wherein the context data is obtained based on at least one of user input for the tagged message or server data for the tagged message, and providing access to the tagged message according to the obtained context data.

Abstract: Techniques are shown for generating image frames from a media presentation, selecting candidate thumbnails from the generated image frames using a selection process, and testing each selected candidate thumbnail for a success ranking relative to a target metric. The probability of choosing a selected candidate thumbnail with a success ranking higher than all other selected thumbnails as an optimum thumbnail for presentation to a user is based, at least in part, on the ratio of the success ranking of the selected candidate thumbnail with the highest success ranking to the sum of the success rankings of all of the selected candidate thumbnails. This Abstract is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims.

Abstract: A method and apparatus for creating and updating a facial image database from a collection of digital images is disclosed. A set of detected faces from a digital image collection is stored in a facial image database, along with data pertaining to them. At least one facial recognition template for each face in the first set is computed, and the images in the set are grouped according to the facial recognition template into similarity groups. Another embodiment is a naming tool for assigning names to a plurality of faces detected in a digital image collection. A facial image database stores data pertaining to facial images detected in images of a digital image collection.

Abstract: A computer-implemented method for automatically sharing media between users is provided. A first image is received depicting a scene with one or more persons. Face detection information is identified for each person in the first image. Identification information for each face detected is obtained without user intervention. The face detection information and the identification information are associated with the first image. The first image including the associated face detection information and identification information is sent to a server hosting a media sharing service. An access link to an event group is received. The event group includes an album of the first image and a first collection of images that share a set of common features with the first image.

Abstract: Biomarkers for predicting the severity of malaria and methods for their detection are disclosed. In one aspect, the present application discloses CXCL4, CXCL10, VEGF, PGDF, IL-1Ra, IL-8, MIP-1?, sFas, Fas-L, sTNF-R2, and sTNF-R1 as biomarkers for the severity of malaria. In another aspect, the present application discloses a method for determining the severity of malaria and predicting mortality due to cerebral malaria. The method comprises the detection of the biomarkers CXCL4 and/or CXCL10 and at least one more biomarker and determining the severity of malaria and predicting mortality due to cerebral malaria based upon the ratio of expression of the biomarkers in the subject versus the expression of the biomarkers in a control.

Abstract: Techniques are shown for generating image frames from a media presentation, selecting candidate thumbnails from the generated image frames using a selection process, and testing each selected candidate thumbnail for a success ranking relative to a target metric. The probability of choosing a selected candidate thumbnail with a success ranking higher than all other selected thumbnails as an optimum thumbnail for presentation to a user is based, at least in part, on the ratio of the success ranking of the selected candidate thumbnail with the highest success ranking to the sum of the success rankings of all of the selected candidate thumbnails. This Abstract is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims.

Abstract: A computer-implemented method for automatically sharing media between users is provided. A first image is received depicting a scene with one or more persons. Face detection information is identified for each person in the first image. Identification information for each face detected is obtained without user intervention. The face detection information and the identification information are associated with the first image. The first image including the associated face detection information and identification information is sent to a server hosting a media sharing service. An access link to an event group is received. The event group includes an album of the first image and a first collection of images that share a set of common features with the first image.

Abstract: A method and apparatus for creating and updating a facial image database from a collection of digital images is disclosed. A set of detected faces from a digital image collection is stored in a facial image database, along with data pertaining to them. At least one facial recognition template for each face in the first set is computed, and the images in the set are grouped according to the facial recognition template into similarity groups. Another embodiment is a naming tool for assigning names to a plurality of faces detected in a digital image collection. A facial image database stores data pertaining to facial images detected in images of a digital image collection.

Abstract: A method and apparatus for creating and updating a facial image database from a collection of digital images is disclosed. A set of detected faces from a digital image collection is stored in a facial image database, along with data pertaining to them. At least one facial recognition template for each face in the first set is computed, and the images in the set are grouped according to the facial recognition template into similarity groups. Another embodiment is a naming tool for assigning names to a plurality of faces detected in a digital image collection. A facial image database stores data pertaining to facial images detected in images of a digital image collection.