Abstract: An adder circuit provides a first operand input and a second operand input to an XNOR cell. The XNOR cell transforms these inputs to a propagate signal that is applied to an OAT cell to produce a carry out signal. A third OAT cell transforms a third operand input and the propagate signal into a sum output signal.

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: Introduced herein is an improved technique of recovering system frequency margin via distributed CPMs. The introduced technique creates and distributes multiple sets of always sensitized critical path replicas across a chip and monitors them for timing failure. The introduced technique takes feedback from these critical path replicas and dynamically boosts the clock frequency of the chip to remove the margin. The introduced technique provides more accurate and more comprehensive coverage of a chip performance.

Abstract: A method involves a headless IoT device wirelessly communicating a MAC address to a client device in response to a scan by the client device, and receiving from the client device a vendor action frame comprising access credentials for communicating via a Wi-Fi access point. The IoT device applies the credentials to authenticate to the Wi-Fi access point, forms an application layer for communicating over the Wi-Fi access point network, and communicates with the client device via the application layer.

Abstract: One embodiment of the present invention includes a parallel processing unit (PPU) that performs pixel shading at variable granularities. For effects that vary at a low frequency across a pixel block, a coarse shading unit performs the associated shading operations on a subset of the pixels in the pixel block. By contrast, for effects that vary at a high frequency across the pixel block, fine shading units perform the associated shading operations on each pixel in the pixel block. Because the PPU implements coarse shading units and fine shading units, the PPU may tune the shading rate per-effect based on the frequency of variation across each pixel group. By contrast, conventional PPUs typically compute all effects per-pixel, performing redundant shading operations for low frequency effects. Consequently, to produce similar image quality, the PPU consumes less power and increases the rendering frame rate compared to a conventional PPU.

Abstract: This disclosure presents a method to denoise a ray traced scene where the ray tracing uses a minimal number of rays. The method can use temporal reprojections to compute a weighted average to the scene data. A spatial filter can be run on the scene data, using the temporal reprojection count to reduce the size of the utilized spatial filter radius. In some aspects, additional temporal filters can be applied to the scene data. In some aspects, global illumination temporal reprojection history counts can be used to modify the spatial filter radius. In some aspects, caustic photon tracing can be conducted to compute a logarithmic cost, which can then be utilized to reduce the denoising radius used by the spatial filter. The modified and adjusted scene data can be sent to a rendering process to complete the rendering to generate a final scene.

Abstract: Multiple snapshots of a scene are captured within an executing application (e.g., a video game). When each snapshot is captured, associated color values per pixel and a distance or depth value z per pixel are stored. The depth information from the snapshots is accessed, and a point cloud representing the depth information is constructed. A mesh structure is constructed from the point cloud. The light field(s) on the surface(s) of the mesh structure are calculated. A surface light field is represented as a texture. A renderer uses the surface light field with geometry information to reproduce the scene captured in the snapshots. The reproduced scene can be manipulated and viewed from different perspectives.

Abstract: Haptic effects have long been provided to enhance content, such as by providing vibrations, rumbles, etc. in a remote controller or other device being used by a user while watching or listening to the content. To date, haptic effects have either been provided by programming controls for the haptic effects within the content itself, or by providing an interface to audio that simply maps certain haptic effects with certain audio frequencies. The present disclosure provides a haptic control interface that intelligently induces haptic effects for content, in particular by using machine learning to detect specific features in content and then induce certain haptic effects for those features.

Abstract: A method, computer readable medium, and system are disclosed for performing traversal stack compression. The method includes traversing a hierarchical data structure having more than two children per node, and during the traversing, creating at least one stack entry, utilizing a processor, where each stack entry contains a plurality of intersected nodes, and adding the at least one stack entry to a compressed traversal stack stored in a memory, utilizing the processor.

Abstract: A style transfer neural network may be used to generate stylized synthetic images, where real images provide the style (e.g., seasons, weather, lighting) for transfer to synthetic images. The stylized synthetic images may then be used to train a recognition neural network. In turn, the trained neural network may be used to predict semantic labels for the real images, providing recognition data for the real images. Finally, the real training dataset (real images and predicted recognition data) and the synthetic training dataset are used by the style transfer neural network to generate stylized synthetic images. The training of the neural network, prediction of recognition data for the real images, and stylizing of the synthetic images may be repeated for a number of iterations. The stylization operation more closely aligns a covariate of the synthetic images to the covariate of the real images, improving accuracy of the recognition neural network.

Abstract: Scene flow represents the three-dimensional (3D) structure and movement of objects in a video sequence in three dimensions from frame-to-frame and is used to track objects and estimate speeds for autonomous driving applications. Scene flow is recovered by a neural network system from a video sequence captured from at least two viewpoints (e.g., cameras), such as a left-eye and right-eye of a viewer. An encoder portion of the system extracts features from frames of the video sequence. The features are input to a first decoder to predict optical flow and a second decoder to predict disparity. The optical flow represents pixel movement in (x,y) and the disparity represents pixel movement in z (depth). When combined, the optical flow and disparity represent the scene flow.

Abstract: Techniques for estimating depth for a video stream captured by a monocular image sensor are disclosed. A sequence of image frames are captured by the monocular image sensor. A first neural network is configured to process at least a portion of the sequence of image frames to generate a depth probability volume. The depth probability volume includes a plurality of probability maps corresponding to a number of discrete depth candidate locations over a range of depths defined for the scene. The depth probability volume can be updated using a second neural network that is configured to generate adaptive gain parameters to integrate the DPVs over time. A third neural network is configured to refine the updated depth probability volume from a lower resolution to a higher resolution that matches the original resolution of the sequence of image frames. A depth map can be calculated based on the depth probability volume.

Abstract: An addressing scheme in systems utilizing a number of operative memory slices in a last level cache that is not evenly divisible by a number of memory channels utilizes the operative slices exposes the full last level cache bandwidth and capacity to data processing logic in a high-performance graphics system.

Abstract: A queue manager apparatus converts inbound commands of a first width into scalar format commands to be queued in a command queue. Furthermore, the queue manager converts the scalar format commands residing in the command queue into outbound commands of a second width for transmission. Converting inbound commands to scalar format commands and then converting the scalar format commands to a target width for transmission allows the queue manager to advantageously provide efficient and programmable command transmission between arbitrary processing units, regardless of potentially mismatched native command widths.

Abstract: In one embodiment, a synchronization instruction causes a processor to ensure that specified threads included within a warp concurrently execute a single subsequent instruction. The specified threads include at least a first thread and a second thread. In operation, the first thread arrives at the synchronization instruction. The processor determines that the second thread has not yet arrived at the synchronization instruction and configures the first thread to stop executing instructions. After issuing at least one instruction for the second thread, the processor determines that all the specified threads have arrived at the synchronization instruction. The processor then causes all the specified threads to execute the subsequent instruction. Advantageously, unlike conventional approaches to synchronizing threads, the synchronization instruction enables the processor to reliably and properly execute code that includes complex control flows and/or instructions that presuppose that threads are converged.

Abstract: Embodiments of the present invention provide a low-latency approach for local or remote application streaming that reaches high FPS targets without overloading the available streaming bandwidth, for example, by limiting the bit rate to the same value that is used by traditional 60 FPS streaming solutions. A client device and server device cooperate to actively monitor and control a video stream to maintain an acceptable balance between latency and video quality by adjusting the frequency or resolution when necessary to improve the streaming experience. When the server device captures and transmits frames at a higher rate, the software stack executing on the client device is able to display frames with less delay, even on a display device limited to 60 Hz, thereby achieving additional latency reduction.

Abstract: Techniques for limiting the growth of errors in decoded data words that arise from bit errors incurred during transmission. The growth of 3+ bit errors in the decoded data word is limited at the expense of a higher number of two bit errors, which are correctable using practical error correcting codes.

Abstract: A technique is applied to eventually converge on a single data storage strategy for any set of object data which had an inconsistent data storage strategy applied during storage while there was a network partition. This state could occur in instances of a highly available distributed object storage system which can store objects according to multiple data storage strategies. Upon the healing of a network partition, the technique discovers if multiple data storage strategies were applied to the object data stored during a network partition, deterministically identifies which data storage strategy represents the correct strategy, for example based on the log of state transitions requested by the client according to the API contract, and ensures that this strategy is consistently applied to all object data in the collection.

Abstract: Head-mounted Displays (HMDs) are commonly used for virtual reality, mixed reality, and augmented reality. HMDs are, by definition, worn on the head of a user to provide a display in the line of sight of the user. By viewing the display, the user is able to experience one of the aforementioned types of reality. Oftentimes, HMDs are configured to integrate live video captured from the user's perspective, especially in the case of the HMD providing augmented reality where a virtual environment is combined with video of the real world. The present disclosure provides a configuration for a HMD having an array of image sensors to accurately capture image data to form the live video from the user's perspective.

Abstract: A neural network structure, namely a warped external recurrent neural network, is disclosed for reconstructing images with synthesized effects. The effects can include motion blur, depth of field reconstruction (e.g., simulating lens effects), and/or anti-aliasing (e.g., removing artifacts caused by sampling frequency). The warped external recurrent neural network is not recurrent at each layer inside the neural network. Instead, the external state output by the final layer of the neural network is warped and provided as a portion of the input to the neural network for the next image in a sequence of images. In contrast, in a conventional recurrent neural network, hidden state generated at each layer is provided as a feedback input to the generating layer. The neural network can be implemented, at least in part, on a processor. In an embodiment, the neural network is implemented on at least one parallel processing unit.