Abstract: A latent code defined in an input space is processed by the mapping neural network to produce an intermediate latent code defined in an intermediate latent space. The intermediate latent code may be used as appearance vector that is processed by the synthesis neural network to generate an image. The appearance vector is a compressed encoding of data, such as video frames including a person's face, audio, and other data. Captured images may be converted into appearance vectors at a local device and transmitted to a remote device using much less bandwidth compared with transmitting the captured images. A synthesis neural network at the remote device reconstructs the images for display.

Abstract: A receiver includes a decision feed forward equalization (DFFE) system coupled to a partial response (PR) system. The partial response system generates, based on a digital signal that includes pre-cursor intersymbol interference (ISI) and post-cursor ISI introduced by a communication channel, a detected signal including a set of detected symbol values. The detected signal is equalized to a partial response. The DFFE system includes a PR inverter to generate a set of estimated transmitted symbol values based on the set of detected symbol values and DFFE circuitry to cancel the pre-cursor ISI and the post-cursor ISI from the detected signal using the set of estimated transmitted symbols and a set of tap coefficients to obtain a compensated signal and a set of compensated symbol values.

Abstract: A method, computer readable medium, and processor are described herein for inline data inspection by using a decoder to decode a load instruction, including a signal to cause a circuit in a processor to indicate whether data loaded by a load instruction exceeds a threshold value. Moreover, an indication of whether data loaded by a load instruction exceeds a threshold value may be stored.

Abstract: In various embodiments, a parallel processor includes a parallel processor module implemented within a first die and a memory system module implemented within a second die. The memory system module is coupled to the parallel processor module via an on-package link. The parallel processor module includes multiple processor cores and multiple cache memories. The memory system module includes a memory controller for accessing a DRAM. Advantageously, the performance of the parallel processor module can be effectively tailored for memory bandwidth demands that typify one or more application domains via the memory system module.

Abstract: Approaches in accordance with various embodiments can perform spatial hash map updates while ensuring the atomicity of the updates for arbitrary data structures. A hash map can be generated for a dataset where entries in the hash map may correspond to multiple independent values, such as pixels of an image to be rendered. Update requests for independent values may be received on multiple concurrent threads, but change requests for independent values corresponding to a hash map entry can be aggregated from a buffer and processed iteratively in a single thread for a given hash map entry. In the case of multi-resolution spatial hashing where data can be stored at various discretization levels, this operation can be repeated to propagate changes from one level to another.

Abstract: Systems and methods enable optimization of a 3D model representation comprising the shape and appearance of a particular 3D scene or object. The opaque 3D mesh (e.g., vertex positions and corresponding topology) and spatially varying material attributes are jointly optimized based on image space losses to match multiple image observations (e.g., reference images of the reference 3D scene or object). A geometric topology defines faces and/or cells in the opaque 3D mesh that are visible and may be randomly initialized and optimized through training based on the image space losses. Applying the geometry topology to an opaque 3D mesh for learning the shape improves accuracy of silhouette edges and performance compared with using transparent mesh representations. In contrast with approaches that require an initial guess for the topology and/or an exhaustive testing of possible geometric topologies, the 3D model representation is learned based on image space differences without requiring an initial guess.

Abstract: Techniques to generate driving scenarios for autonomous vehicles characterize a path in a driving scenario according to metrics such as narrowness and effort. Nodes of the path are assigned a time for action to avoid collision from the node. The generated scenarios may be simulated in a computer.

Abstract: A technique for block data transfer is disclosed that reduces data transfer and memory access overheads and significantly reduces multiprocessor activity and energy consumption. Threads executing on a multiprocessor needing data stored in global memory can request and store the needed data in on-chip shared memory, which can be accessed by the threads multiple times. The data can be loaded from global memory and stored in shared memory using an instruction which directs the data into the shared memory without storing the data in registers and/or cache memory of the multiprocessor during the data transfer.

Abstract: Various examples of the present disclosure include a stereoscopic deep neural network (DNN) that produces accurate and reliable results in real-time. Both LIDAR data (supervised training) and photometric error (unsupervised training) may be used to train the DNN in a semi-supervised manner. The stereoscopic DNN may use an exponential linear unit (ELU) activation function to increase processing speeds, as well as a machine learned argmax function that may include a plurality of convolutional layers having trainable parameters to account for context. The stereoscopic DNN may further include layers having an encoder/decoder architecture, where the encoder portion of the layers may include a combination of three-dimensional convolutional layers followed by two-dimensional convolutional layers.

Abstract: In various examples, systems and methods are disclosed that preserve rich spatial information from an input resolution of a machine learning model to regress on lines in an input image. The machine learning model may be trained to predict, in deployment, distances for each pixel of the input image at an input resolution to a line pixel determined to correspond to a line in the input image. The machine learning model may further be trained to predict angles and label classes of the line. An embedding algorithm may be used to train the machine learning model to predict clusters of line pixels that each correspond to a respective line in the input image. In deployment, the predictions of the machine learning model may be used as an aid for understanding the surrounding environment—e.g., for updating a world model—in a variety of autonomous machine applications.

Abstract: Described approaches provide for effectively and scalably using multiple GPUs to build and probe hash tables and materialize results of probes. Random memory accesses by the GPUs to build and/or probe a hash table may be distributed across GPUs and executed concurrently using global location identifiers. A global location identifier may be computed from data of an entry and identify a global location for an insertion and/or probe using the entry. The global location identifier may be used by a GPU to determine whether to perform an insertion or probe using an entry and/or where the insertion or probe is to be performed. To coordinate GPUs in materializing results of probing a hash table a global offset to the global output buffer may be maintained in memory accessible to each of the GPUs or the GPUs may compute global offsets using an exclusive sum of the local output buffer sizes.

Abstract: In various examples, a current claimed set of points representative of a volume in an environment occupied by a vehicle at a time may be determined. A vehicle-occupied trajectory and at least one object-occupied trajectory may be generated at the time. An intersection between the vehicle-occupied trajectory and an object-occupied trajectory may be determined based at least in part on comparing the vehicle-occupied trajectory to the object-occupied trajectory. Based on the intersection, the vehicle may then execute the first safety procedure or an alternative procedure that, when implemented by the vehicle when the object implements the second safety procedure, is determined to have a lesser likelihood of incurring a collision between the vehicle and the object than the first safety procedure.

Abstract: A style-based generative network architecture enables scale-specific control of synthesized output data, such as images. During training, the style-based generative neural network (generator neural network) includes a mapping network and a synthesis network. During prediction, the mapping network may be omitted, replicated, or evaluated several times. The synthesis network may be used to generate highly varied, high-quality output data with a wide variety of attributes. For example, when used to generate images of people's faces, the attributes that may vary are age, ethnicity, camera viewpoint, pose, face shape, eyeglasses, colors (eyes, hair, etc.), hair style, lighting, background, etc. Depending on the task, generated output data may include images, audio, video, three-dimensional (3D) objects, text, etc.

Abstract: A style-based generative network architecture enables scale-specific control of synthesized output data, such as images. During training, the style-based generative neural network (generator neural network) includes a mapping network and a synthesis network. During prediction, the mapping network may be omitted, replicated, or evaluated several times. The synthesis network may be used to generate highly varied, high-quality output data with a wide variety of attributes. For example, when used to generate images of people's faces, the attributes that may vary are age, ethnicity, camera viewpoint, pose, face shape, eyeglasses, colors (eyes, hair, etc.), hair style, lighting, background, etc. Depending on the task, generated output data may include images, audio, video, three-dimensional (3D) objects, text, etc.

Abstract: Systems and methods of presenting interrupting content during human speech are disclosed. The proposed systems offer improved duplex communications in conversational AI platforms. In some embodiments, the system receives speech data and evaluates the data using linguistic models. If the linguistic models detect indications of linguistic irregularities such as mispronunciation, a smart feedback assistant can determine that the system should interrupt the speaker in near-real-time and provide feedback regarding their pronunciation. In addition, conversational irregularities may also be detected, causing the smart feedback assistant to interrupt with presentation of moderating guidance. In some cases, emotion models may also be utilized to detect emotional states based on the speaker's voice in order to offer near-immediate feedback. Users can also customize the manner and occasions in which they are interrupted.

Abstract: In examples, a filter used to denoise shadows for a pixel(s) may be adapted based at least on variance in temporally accumulated ray-traced samples. A range of filter values for a spatiotemporal filter may be defined based on the variance and used to exclude temporal ray-traced samples that are outside of the range. Data used to compute a first moment of a distribution used to compute variance may be used to compute a second moment of the distribution. For binary signals, such as visibility, the first moment (e.g., accumulated mean) may be equivalent to a second moment (e.g., the mean squared). In further respects, spatial filtering of a pixel(s) may be skipped based on comparing the mean of variance of the pixel(s) to one or more thresholds and based on the accumulated number of values for the pixel.

Abstract: A device including a stack of dies. Each of the dies can have unit stair-step conductive paths of connection features which include through-die via structures and routing structures. The unit stair-step conductive paths of one of the dies can be interconnected to another one of the unit stair-step conductive paths of another one of the dies to form one of a plurality conductive stair-case structures through two or more of the dies. The unit stair-step conductive paths can be connected to reduce signal cross talk between the conductive stair-case structures whereby at least some of the conductive stair-case structures are connected to transmit a same polarity of electrical signals are spatially separated in a dimension that is perpendicular to a major surface of the dies. A method of manufacturing the device is also disclosed.

Abstract: A robotic control system directs a robot to take an object from a human grasp by obtaining an image of a human hand holding an object, estimating the pose of the human hand and the object, and determining a grasp pose for the robot that will not interfere with the human hand. In at least one example, a depth camera is used to obtain a point cloud of the human hand holding the object. The point cloud is provided to a deep network that is trained to generate a grasp pose for a robotic gripper that can take the object from the human's hand without pinching or touching the human's fingers.

Abstract: According to an aspect of an embodiment, operations may comprise receiving, from a LIDAR mounted on a vehicle, a first 3D point cloud comprising points of a region around the vehicle as observed by the LIDAR. The operations may also comprise accessing an HD map comprising a second 3D point cloud comprising points of the region around the vehicle. The operations may also comprise segmenting LIDAR ground points from LIDAR non-ground points in the first 3D point cloud. The operations may also comprise segmenting map ground points from map non-ground points in the second 3D point cloud. The operations may also comprise determining a pose of the vehicle by matching the LIDAR ground points to the map ground points and by matching the LIDAR non-ground points to the map non-ground points.

Abstract: A neural network is trained to identify one or more features of an image. The neural network is trained using a small number of original images, from which a plurality of additional images are derived. The additional images generated by rotating and decoding embeddings of the image in a latent space generated by an autoencoder. The images generated by the rotation and decoding exhibit changes to a feature that is in proportion to the amount of rotation.