Abstract: In various examples, at least partial control of a vehicle may be transferred to a control system remote from the vehicle. Sensor data may be received from a sensor(s) of the vehicle and the sensor data may be encoded to generate encoded sensor data. The encoded sensor data may be transmitted to the control system for display on a virtual reality headset of the control system. Control data may be received by the vehicle and from the control system that may be representative of a control input(s) from the control system, and actuation by an actuation component(s) of the vehicle may be caused based on the control input.

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: The technology disclosed herein involves tracking contention and using the tracked contention to reduce latency of exclusive memory operations. The technology enables a processor to track which locations in main memory are contentious and to modify the order exclusive memory operations are processed based on the contentiousness. A thread can include multiple exclusive operations for the same memory location (e.g., exclusive load and a complementary exclusive store). The multiple exclusive memory operations can be added to a queue and include one or more intervening operations between them in the queue. The processor may process the operations in the queue based on the order they were added and may use the tracked contention to perform out-of-order processing for some of the exclusive operations. For example, the processor can execute the exclusive load operation and because the corresponding location is contentious can process the complementary exclusive store operation before the intervening operations.

Abstract: An end-to-end system for data generation, map creation using the generated data, and localization to the created map is disclosed. Mapstreams—or streams of sensor data, perception outputs from deep neural networks (DNNs), and/or relative trajectory data—corresponding to any number of drives by any number of vehicles may be generated and uploaded to the cloud. The mapstreams may be used to generate map data—and ultimately a fused high definition (HD) map—that represents data generated over a plurality of drives. When localizing to the fused HD map, individual localization results may be generated based on comparisons of real-time data from a sensor modality to map data corresponding to the same sensor modality. This process may be repeated for any number of sensor modalities and the results may be fused together to determine a final fused localization result.

Abstract: A three-dimensional (3D) density volume of an object is constructed from tomography images (e.g., x-ray images) of the object. The tomography images are projection images that capture all structures of an object (e.g., human body) between a beam source and imaging sensor. The beam effectively integrates along a path through the object producing a tomography image at the imaging sensor, where each pixel represents attenuation. A 3D reconstruction pipeline includes a first neural network model, a fixed function backprojection unit, and a second neural network model. Given information for the capture environment, the tomography images are processed by the reconstruction pipeline to produce a reconstructed 3D density volume of the object. In contrast with a set of 2D slices, the entire 3D density volume is reconstructed, so two-dimensional (2D) density images may be produced by slicing through any portion of the 3D density volume at any angle.

Abstract: Disclosed approaches provide for irradiance caches which may be used to share irradiance between ray interactions spatially and/or temporally. An irradiance cache may store incoming irradiance or outgoing irradiance and may be updated by casting one or more rays from one or more locations to sample irradiance for the location(s). The number of rays that are cast may be reduced by ranking the locations, irradiance caches, and/or corresponding groups of geometry based on one or more characteristics thereof. For example, a ranking score may be computed based on camera distance, camera visibility, and/or a number of frames since a prior update. When sampling a location, outgoing irradiance from an outgoing irradiance cache may be used to determine shading when a hit distance of a ray used to generate the sample exceeds a threshold value.

Abstract: A hardware-based traversal coprocessor provides acceleration of tree traversal operations searching for intersections between primitives represented in a tree data structure and a ray. The primitives may include opaque and alpha triangles used in generating a virtual scene. The hardware-based traversal coprocessor is configured to determine primitives intersected by the ray, and return intersection information to a streaming multiprocessor for further processing. The hardware-based traversal coprocessor is configured to provide a deterministic result of intersected triangles regardless of the order that the memory subsystem returns triangle range blocks for processing, while opportunistically eliminating alpha intersections that lie further along the length of the ray than closer opaque intersections.

Abstract: Apparatuses, systems, and techniques to decode encoded data. In at least one embodiment, parts of information for decoding the encoded data is provided to a plurality of processors, and parts of data decoded by the plurality of processors is combined.

Abstract: In various examples, sensor data representative of a field of view of at least one sensor of a vehicle in an environment is received from the at least one sensor. Based at least in part on the sensor data, parameters of an object located in the environment are determined. Trajectories of the object are modeled toward target positions based at least in part on the parameters of the object. From the trajectories, safe time intervals (and/or safe arrival times) over which the vehicle occupying the plurality of target positions would not result in a collision with the object are computed. Based at least in part on the safe time intervals (and/or safe arrival times) and a position of the vehicle in the environment a trajectory for the vehicle may be generated and/or analyzed.

Abstract: Edge-connected semiconductor systems are described along with methods of making and using the same. First and second integrated circuit packages are obtained, each including a substrate assembly having top and bottom sides and an edge that extends between the top and the bottom sides. Edge contacts are disposed on the edges of the substrate assemblies. A ganged assembly is formed by establishing conductive paths between the edge contacts of the substrate assemblies. The ganged assembly is coupled to a printed circuit board (“PCB”) by coupling host contacts on one or more of the substrate assemblies to corresponding contacts on the PCB.

Abstract: In various examples, a deep neural network (DNN) is trained to accurately predict, in deployment, distances to objects and obstacles using image data alone. The DNN may be trained with ground truth data that is generated and encoded using sensor data from any number of depth predicting sensors, such as, without limitation, RADAR sensors, LIDAR sensors, and/or SONAR sensors. Camera adaptation algorithms may be used in various embodiments to adapt the DNN for use with image data generated by cameras with varying parameters—such as varying fields of view. In some examples, a post-processing safety bounds operation may be executed on the predictions of the DNN to ensure that the predictions fall within a safety-permissible range.

Abstract: Data bits are encoded in an eleven bit seven pulse amplitude modulated three-level (PAM-3) symbol format on a plurality of data channels and two auxiliary data channels, and one or more of a cyclic redundancy check (CRC) value, a poison value, and a severity value are encoded as PAM-3 symbols on an error correction channel.

Abstract: In various examples, object detections of a machine learning model are leveraged to automatically generate new ground truth data for images captured at different perspectives. The machine learning model may generate a prediction of a detected object at the different perspective, and an object tracking algorithm may be used to track the object through other images in a sequence of images where the machine learning model may not have detected the object. New ground truth data may be generated as a result of the object tracking algorithms outputs, and the new ground truth data may be used to retrain or update the machine learning model, train a different machine learning model, or increase the robustness of a ground truth data set that may be used for training machine learning models from various perspectives.

Abstract: A system includes a first device and a second device coupled to a link including two or more data paths and a first portion and a second portion. The first device is to transmit a number of bits corresponding to a message before training the link on the first portion of the link, where the number of bits is equal to a number of the two or more data paths, and where each data path transmits one bit of the number of bits. The second device is to receive the message before training the link. The second device is to perform a decode operation on the number of bits received to determine the corresponding message and transmit a second message or data on the second portion of the link in response to performing the decode operation on the number of bits.

Abstract: A target image corresponding to a novel view may be synthesized from two source images, corresponding source camera poses, and pixel attribute correspondences between the two source images. A particular object in the target image need only be visible in one of the two source images for successful synthesis. Each pixel in the target image is defined according to an identified pixel in one of the two source images. The identified source pixel provides attributes such as color, texture, and feature descriptors for the target pixel. The source and target camera poses are used to define geometric relationships for identifying the source pixels. In an embodiment, the pixel attribute correspondences are optical flow that defines movement of attributes from a first image of the two source images to a second image of the two source images.

Abstract: Apparatuses, systems, and techniques are presented to generate image or video content representing at least one point of view. In at least one embodiment, one or more neural networks are used to generate one or more images of one or more objects from a first point of view based at least in part upon one or more images of the one or more objects from a second point of view.

Abstract: Approaches presented herein can reduce temporal lag that may be introduced in a generated image sequence that utilizes temporal accumulation for denoising in dynamic scenes. A fast historical frame can be generated along with a full historical frame generated for a denoising process, with the fast historical frame being accumulated using an exponential moving average with a significantly higher blend weight. This fast history frame can be used to determine a clamping window that can be used to clamp a corresponding full historical value before, or after, reprojection. The fast historical blend weight can be adjusted to control the amount of noise versus temporal lag in an image sequence. In some embodiments, differences between fast and full historical values can also be used to determine an amount of spatial filtering to be applied.

Abstract: A circuit includes a set of multiple bit generating cells. One or more adjustable characterization circuits are coupled to inputs to the bit generating cells to affect the outputs of the bit generating cells. Based on the effect of the characterization circuit(s) on the outputs of the bit generating cells, a subset less than all of the bit generating cells is selected.

Abstract: A display device for measuring the end-to-end latency of a computing system. The computing system includes an input device, a computing device, and the display device. The display device is directly connected with the input device and receives input data packets generated by the input device in response to received user input events. The display device passes the input packets to the computing device for graphics processing. The display device measures the end-to-end latency comprising the sum of three latencies. A first latency comprises an input delay of the input device. A second latency comprises an amount of time between generation of the input packet and a corresponding change in pixel values caused by the input event at the display device. A third latency comprises a display latency. The display device also displays latency information associated with the measured end-to-end latency.

Abstract: According to an aspect of an embodiment, operations may comprise accessing high definition (HD) map data of a region, presenting, via a user interface, information describing the HD map data, receiving instructions, via the user interface, for modifying the HD map data by adding one or more synthetic objects to locations in the HD map data, modifying the HD map data based on the received instructions, and generating a synthetic track in the modified HD map data comprising, for each of one or more vehicle poses, generated synthetic sensor data based on the one or more synthetic objects in the modified HD map data.