Abstract: One embodiment provides a method comprising receiving a piece of content and salient moments data for the piece of content. The method further comprises, based on the salient moments data, determining a first path for a viewport for the piece of content. The method further comprises displaying the viewport on a display device. Movement of the viewport is based on the first path during playback of the piece of content. The method further comprises generating an augmentation for a salient moment occurring in the piece of content, and presenting the augmentation in the viewport during a portion of the playback. The augmentation comprises an interactive hint for guiding the viewport to the salient moment.

Abstract: One embodiment provides a method comprising receiving a first input image and a second input image. Each input image comprises a facial image of an individual. For each input image, a first set of facial regions of the facial image is distinguished from a second set of facial regions of the facial image based on a learning based model. The first set of facial regions comprises age-invariant facial features, and the second set of facial regions comprises age-sensitive facial features. The method further comprises determining whether the first input image and the second input images comprise facial images of the same individual by performing face verification based on the first set of facial regions of each input image.

Abstract: One embodiment provides a method comprising receiving an input, and classifying the input utilizing a learned linear combination of multi-dimensional filters. Each multi-dimensional filter identifies a multi-dimensional pattern of a homogenous feature. The method further comprises generating an output indicative of a classification of the input.

Abstract: One embodiment provides a method comprising receiving a first input image and a second input image. Each input image comprises a facial image of an individual. For each input image, a first set of facial regions of the facial image is distinguished from a second set of facial regions of the facial image based on a learning based model. The first set of facial regions comprises age-invariant facial features, and the second set of facial regions comprises age-sensitive facial features. The method further comprises determining whether the first input image and the second input images comprise facial images of the same individual by performing face verification based on the first set of facial regions of each input image.

Abstract: One embodiment provides a method comprising receiving a piece of content and salient moments data for the piece of content. The method further comprises, based on the salient moments data, determining a first path for a viewport for the piece of content. The method further comprises displaying the viewport on a display device. Movement of the viewport is based on the first path during playback of the piece of content. The method further comprises generating an augmentation for a salient moment occurring in the piece of content, and presenting the augmentation in the viewport during a portion of the playback. The augmentation comprises an interactive hint for guiding the viewport to the salient moment.

Abstract: One embodiment provides a method comprising receiving an input, and classifying the input utilizing a learned linear combination of multi-dimensional filters. Each multi-dimensional filter identifies a multi-dimensional pattern of a homogenous feature. The method further comprises generating an output indicative of a classification of the input.

Abstract: Various embodiments provide techniques and constructs to improve execution speed of distributed iterative computation using heterogeneous specialized resources including, for example, processors and accelerators. Iteration over an arbitrary sub-graph without loop unrolling including for algorithms with data-dependent loop termination and large iteration counts, including as a result of nested iteration, are supported in a resource-efficient manner without adding vertices to a dataflow graph to represent iteration constructs. Instead, some or all of the existing vertices within the sub-graph that is to be iterated upon based on having additional and/or modified ports and channels associated with them.

Abstract: Various embodiments provide techniques and constructs to improve execution speed of distributed iterative computation using heterogeneous specialized resources including, for example, processors and accelerators. Iteration over an arbitrary sub-graph without loop unrolling including for algorithms with data-dependent loop termination and large iteration counts, including as a result of nested iteration, are supported in a resource-efficient manner without adding vertices to a dataflow graph to represent iteration constructs. Instead, some or all of the existing vertices within the sub-graph that is to be iterated upon based on having additional and/or modified ports and channels associated with them.

Abstract: A connection bridge mechanism for adding arbitrary protocols into a middleware-based application without accessing the source code for the middleware or producing a new version of the middleware source code. Using a connection bridge allows the plugged in protocol to be utilized by the middleware without requiring that the same O/S primitives which are utilized internally by the middleware be supported by the plugged in protocol. The plugged in protocol can be utilized in addition to the internally supported middleware protocols or it can replace the existing middleware protocols. The connection bridge provides the communication between the protocols and the middleware package. The connection bridge queues the requests from the protocols and notifies the middleware when there are requests pending for action by the middleware.

Abstract: A logging system which accepts logged messages concurrently from any number of parts of the application software. Each message is written to a message queue and then read and passed to a message forwarder, which the message forwarder analyzes one or more of the data fields of the message in order to decide whether to forward a particular message. In the event that the message will be forwarded, the forwarder also determines which destination or destinations to forward it to, which data fields from the logged message will be forwarded, and how those fields will be formatted when forwarded. The forwarder code executes on its own Operating System thread or task to decouple the continued execution of the application from the process of forwarding the messages to their destination or destinations.