Abstract: A method for remotely provisioning resources for running a computer application is described. The method includes: causing, using one or more processing units, an initialization of a user interactive video portion of a computer application, the computer application being executed using a remote server; determining a runtime of a static video portion of the computer application and a time required to complete initialization of the user interactive portion using information provided by the remote server; and delaying a start time of displaying the static video portion when the runtime of the static video portion is shorter than the time required to complete the initialization of the user interactive portion. A device that is capable of performing the above method and a server are also described.

Abstract: Systems and methods for evaluating cooling characteristics are disclosed. In at least one embodiment, a thermal test vehicle can be used to simulate a thermal performance of a computing unit.

Abstract: Various embodiments include a memory device that is capable of transferring both commands and data via a single clock signal input. In order to initialize the memory device to receive commands, a memory controller transmits a synchronization command to the memory device. The synchronization command establishes command start points that identify the beginning clock cycle of a command that is transferred to the memory device over multiple clock cycles. Thereafter, the memory controller transmits subsequent commands to the memory device according to a predetermined command length. The predetermined command length is based on the number of clock cycles needed to transfer each command to the memory device. Adjacent command start points are separated from one another by the predetermined command length. In this manner, the memory device avoids the need for a second lower speed clock signal for transferring commands to the memory device.

Abstract: A method, computer readable medium, and system are disclosed for overlaying a cell onto a polygon meshlet. The polygon meshlet may include a grouping of multiple geometric shapes such as triangles, and the cell may include a square-shaped boundary. Additionally, every polygon (e.g., a triangle or other geometric shape) within the polygon meshlet that has at least one edge fully inside the cell is removed to create an intermediate meshlet. A selected vertex is determined from all vertices (e.g., line intersections) of the intermediate meshlet that are located within the cell, based on one or more criteria, and all the vertices of the intermediate meshlet that are located within the cell are replaced with the selected vertex to create a modified meshlet. The modified meshlet is then rendered (e.g., as part of a process to generate a scene to be viewed).

Abstract: In various examples, a sequential deep neural network (DNN) may be trained using ground truth data generated by correlating (e.g., by cross-sensor fusion) sensor data with image data representative of a sequences of images. In deployment, the sequential DNN may leverage the sensor correlation to compute various predictions using image data alone. The predictions may include velocities, in world space, of objects in fields of view of an ego-vehicle, current and future locations of the objects in image space, and/or a time-to-collision (TTC) between the objects and the ego-vehicle. These predictions may be used as part of a perception system for understanding and reacting to a current physical environment of the ego-vehicle.

Abstract: One embodiment of a computer-implemented method for compiling a material graph into a set of instructions for execution within an execution unit includes receiving a first material graph having a plurality of nodes, wherein each node included in the plurality of nodes represents a different surface property of a material; parsing the material graph to generate an expression tree that includes one or more expressions for each node included in the plurality of nodes; and generating a set of byte code instructions corresponding to the material graph based on the expression tree, wherein the byte code instructions are executable by a plurality of processing cores included within the execution unit.

Abstract: Apparatuses, systems, and techniques for isolating the performance of a quality-of-service (QoS) policy for improved data streaming systems and applications. In at least one embodiment, a metric is determined for a QoS policy used to provide an application session based on a value of at least one characteristic of the application session that reflects an impact of one or more external conditions beyond the control of the QoS policy.

Abstract: In various examples, systems and methods are disclosed that accurately identify driver and passenger in-cabin activities that may indicate a biomechanical distraction that prevents a driver from being fully engaged in driving a vehicle. In particular, image data representative of an image of an occupant of a vehicle may be applied to one or more deep neural networks (DNNs). Using the DNNs, data indicative of key point locations corresponding to the occupant may be computed, a shape and/or a volume corresponding to the occupant may be reconstructed, a position and size of the occupant may be estimated, hand gesture activities may be classified, and/or body postures or poses may be classified. These determinations may be used to determine operations or settings for the vehicle to increase not only the safety of the occupants, but also of surrounding motorists, bicyclists, and pedestrians.

Abstract: A network device serving as a local VXLAN) Tunnel Endpoint (VTEP) includes a communication interface, a first processor and a packet processor. The communication interface communicates between the local VTEP and remote VTEPs, each VTEP has a respective VXLAN Identifier (VNI). The first processor imports a Downstream-VNI (D-VNI) to be used in forwarding packets from the local VTEP to a remote VTEP, creates a unique egress Routing Interface (RIF) that is translatable into the imported D-VNI, and associates the unique egress RIF with one or more route entries in the local VTEP. The packet processor receives a packet destined to the remote VTEP, looks up the packet in the route entries in the local VTEP to retrieve the unique egress RIF, translates the unique egress RIF into the imported D-VNI, encapsulates the packet with the imported D-VNI, and forwards the encapsulated packet in accordance with the unique egress RIF.

Abstract: In an embodiment, an augmented reality display is provided that incorporates a prescription lens for the wearer. In an embodiment, an image is generated from a display and directed into the edge of the prescription lens, and the lens acts as a waveguide. The image is internally reflected within the prescription lens, and is directed to the wearer by an image combiner embedded within the prescription lens. In an embodiment, the augmented reality display can be adjusted for many common vision problems including myopia, hyperopia, astigmatism, and presbyopia.

Abstract: Techniques are disclosed for improving the throughput of ray intersection or visibility queries performed by a ray tracing hardware accelerator. Throughput is improved, for example, by releasing allocated resources before ray visibility query results are reported by the hardware accelerator. The allocated resources are released when the ray visibility query results can be stored in a compressed format outside of the allocated resources. When reporting the ray visibility query results, the results are reconstructed based on the results stored in the compressed format. The compressed format storage can be used for ray visibility queries that return no intersections or terminate on any hit ray visibility query. One or more individual components of allocated resources can also be independently deallocated based on the type of data to be returned and/or results of the ray visibility query.

Abstract: In various examples, systems and methods are disclosed that accurately identify driver and passenger in-cabin activities that may indicate a biomechanical distraction that prevents a driver from being fully engaged in driving a vehicle. In particular, image data representative of an image of an occupant of a vehicle may be applied to one or more deep neural networks (DNNs). Using the DNNs, data indicative of key point locations corresponding to the occupant may be computed, a shape and/or a volume corresponding to the occupant may be reconstructed, a position and size of the occupant may be estimated, hand gesture activities may be classified, and/or body postures or poses may be classified. These determinations may be used to determine operations or settings for the vehicle to increase not only the safety of the occupants, but also of surrounding motorists, bicyclists, and pedestrians.

Abstract: A ray (e.g., a traced path of light, etc.) is generated from an originating pixel within a scene being rendered. Additionally, one or more shadow map lookups are performed for the originating pixel to estimate an intersection of the ray with alpha-tested geometry within the scene. A shadow map stores the distance of geometry as seen from the point of view of the light, and alpha-tested geometry includes objects within the scene being rendered that have a determined texture and opacity. Further, the one or more shadow map lookups are performed to determine a visibility value for the pixel (e.g., that identifies whether the originating pixel is in a shadow) and a distance value for the pixel (e.g., that identifies how far the pixel is from the light). Further still, the visibility value and the distance value for the pixel are passed to a denoiser.

Abstract: Apparatuses, systems, and techniques are presented to reduce temporal lag when a dynamic event is occurring in computer generated video. In one embodiment, a first averaging algorithm is utilized to determine a display value for a pixel based at least in part on previous pixel values. Once a dynamic event is detected, a set of the previous pixel values is averaged using a second averaging algorithm. The pixel value is updated based on the first averaging of the pixel values and the second averaging of the pixel values to determine a current pixel value.

Abstract: Apparatuses, systems, and techniques to train a generative model based at least in part on a private dataset. In at least one embodiment, the generative model is trained based at least in part on a differentially private Sinkhorn algorithm, for example, using backpropagation with gradient descent to determine a gradient of a set of parameters of the generative models and modifying the set of parameters based at least in part on the gradient.

Abstract: A method, computer program product, and system perform computations using a processor. A first instruction including a first index vector operand and a second index vector operand is received and the first index vector operand is decoded to produce first coordinate sets for a first array, each first coordinate set including at least a first coordinate and a second coordinate of a position of a non-zero element in the first array. The second index vector operand is decoded to produce second coordinate sets for a second array, each second coordinate set including at least a third coordinate and a fourth coordinate of a position of a non-zero element in the second array. The first coordinate sets are summed with the second coordinate sets to produce output coordinate sets and the output coordinate sets are converted into a set of linear indices.

Abstract: A neural network is trained to focus on a domain of interest. For example, in a pre-training phase, the neural network in trained using synthetic training data, which is configured to omit or limit content less relevant to the domain of interest, by updating parameters of the neural network to improve the accuracy of predictions. In a subsequent training phase, the pre-trained neural network is trained using real-world training data by updating only a first subset of the parameters associated with feature extraction, while a second subset of the parameters more associated with policies remains fixed.

Abstract: In various examples, systems and methods are disclosed that detect hazards on a roadway by identifying discontinuities between pixels on a depth map. For example, two synchronized stereo cameras mounted on an ego-machine may generate images that may be used extract depth or disparity information. Because a hazard's height may cause an occlusion of the driving surface behind the hazard from a perspective of a camera(s), a discontinuity in disparity values may indicate the presence of a hazard. For example, the system may analyze pairs of pixels on the depth map and, when the system determines that a disparity between a pair of pixels satisfies a disparity threshold, the system may identify the pixel nearest the ego-machine as a hazard pixel.

Abstract: In various examples, locations of directional landmarks, such as vertical landmarks, may be identified using 3D reconstruction. A set of observations of directional landmarks (e.g., images captured from a moving vehicle) may be reduced to 1D lookups by rectifying the observations to align directional landmarks along a particular direction of the observations. Object detection may be applied, and corresponding 1D lookups may be generated to represent the presence of a detected vertical landmark in an image.

Abstract: In training a deep neural network using reduced precision, gradient computation operates on larger values without affecting the rest of the training procedure. One technique trains the deep neural network to develop loss, scales the loss, computes gradients at a reduced precision, and reduces the magnitude of the computed gradients to compensate for scaling of the loss. In one example non-limiting arrangement, the training forward pass scales a loss value by some factor S and the weight update reduces the weight gradient contribution by 1/S. Several techniques can be used for selecting scaling factor S and adjusting the weight update.