Abstract: A computer, including a processor and a memory, the memory including instructions to be executed by the processor to detect a moving object in video stream data based on determining an eccentricity map. The instructions can further include instructions to determine a magnitude and direction of motion of the moving object, transform the magnitude and direction to global coordinates and operate a vehicle based on the transformed magnitude and direction.

Abstract: A computing system can crop an image based on a width, height and location of a first vehicle in the image. The computing system can estimate a pose of the first vehicle based on inputting the cropped image and the width, height and location of the first vehicle into a deep neural network. The computing system can then operate a second vehicle based on the estimated pose. The computing system may train a model to identify a type and a location of a hazard according to the estimated pose, the hazard being such things as ice, mud, pothole, or other hazard. The model may be used by an autonomous vehicle to identify and avoid hazards or to provide drive assistance alerts.

Abstract: A system, comprising a computer that includes a processor and a memory, the memory storing instructions executable by the processor to input a red-green-blue (RGB) image and an eccentricity image to a neural network which outputs a located object based on combining the RGB image and the eccentricity image, wherein the eccentricity image is based on a per-pixel rolling average and a per-pixel rolling variance over a moving window of k video frames. The memory can further include instructions executable by the processor to receive the located object at a computing device included in one or more of a vehicle or a traffic information system.

Abstract: In a message receiver, a receipt timestamp is assigned according to a time of wirelessly receiving a message. Object data about an object is extracted from the message, including a timestamp adjustment for the object from a message sender. An estimated sender capture timestamp is assigned to the object by subtracting the timestamp adjustment and an estimated message transmission latency from the receipt timestamp; a receiver capture timestamp is assigned to receiver object data captured in the receiver. The sender object data and the receiver object data are combined according to the estimated sender capture timestamp and the receiver capture timestamp, thereby generating combined object data.

Abstract: A system, comprising a computer that includes a processor and a memory, the memory storing instructions executable by the processor to determine an object location prediction based on a video data stream, wherein the object location prediction is based on processing cropped TEDA data with a neural network. The processor can be further programmed to download the object location prediction to a vehicle based on a location of the vehicle.

Abstract: A computing system can receive an image including foreground pixels. The foreground pixels can be determined based on determining eccentricity ?k based on a sequence of images acquired by a stationary sensor. The vehicle can determine moving objects in the image based on the foreground pixels. The vehicle can be operated based on the moving objects in the image.

Abstract: The present invention extends to methods, systems, and computer program products for refining synthetic data with a Generative Adversarial Network (GAN) using auxiliary inputs. Refined synthetic data can be rendered more realistically than the original synthetic data. Refined synthetic data also retains annotation metadata and labeling metadata used for training of machine learning models. GANs can be extended to use auxiliary channels as inputs to a refiner network to provide hints about increasing the realism of synthetic data. Refinement of synthetic data enhances the use of synthetic data for additional applications.

Abstract: A computing system can crop an image based on a width, height and location of a first vehicle in the image. The computing system can estimate a pose of the first vehicle based on inputting the cropped image and the width, height and location of the first vehicle into a deep neural network. The computing system can then operate a second vehicle based on the estimated pose. The computing system may train a model to identify a type and a location of a hazard according to the estimated pose, the hazard being such things as ice, mud, pothole, or other hazard. The model may be used by an autonomous vehicle to identify and avoid hazards or to provide drive assistance alerts.

Abstract: A system and methods are described for generating a super-resolution depth-map. A method includes: determining a plurality of unmeasured depth-map positions using measured depth-elements from a first sensor and spatial-elements from a second sensor; for each of the plurality, calculating estimated depth-elements using a gradient-based optimization; and generating a super-resolution depth-map that comprises the measured and estimated depth-elements.

Abstract: A computing system can receive, in a vehicle, moving object information is determined by processing lidar sensor data acquired by a stationary lidar sensor. The moving object information can be determined using typicality and eccentricity data analysis (TEDA) on the lidar sensor data. The vehicle can be operated based on the moving object information.

Abstract: A computing system can receive a stabilized image stream, wherein the stabilized image stream is drift-corrected based on determining that an input image stream is stable and then applying drift correction to maintain a stabilized field of view, wherein the field of view is stabilized with respect to the real world. The computing system can operate a vehicle based on determining at least one moving object in the stabilized image stream.

Abstract: A high-level vehicle command is determined based on a location of the vehicle with respect to a route including a start location and a finish location. An image is acquired of the vehicle external environment. Steering, braking, and powertrain commands are determined based on inputting the high-level command and the image into a Deep Neural Network. The vehicle is operated by actuating vehicle components based on the steering, braking and powertrain commands.

Abstract: A computing system can determine moving objects in a sequence of images based on recursively calculating red-green-blue (RGB) eccentricity 249 k based on a video data stream. A vehicle can be operated based on the determined moving objects. The video data stream can be acquired by a color video sensor included in the vehicle or a traffic infrastructure system.

Abstract: A computing system can receive an image including foreground pixels. The foreground pixels can be determined based on determining eccentricity ?k based on a sequence of images acquired by a stationary sensor. The vehicle can determine moving objects in the image based on the foreground pixels. The vehicle can be operated based on the moving objects in the image.

Abstract: The present invention extends to methods, systems, and computer program products for classifying time series image data. Aspects of the invention include encoding motion information from video frames in an eccentricity map. An eccentricity map is essentially a static image that aggregates apparent motion of objects, surfaces, and edges, from a plurality of video frames. In general, eccentricity reflects how different a data point is from the past readings of the same set of variables. Neural networks can be trained to detect and classify actions in videos from eccentricity maps. Eccentricity maps can be provided to a neural network as input. Output from the neural network can indicate if detected motion in a video is or is not classified as an action, such as, for example, a hand gesture.

Abstract: A computing system can receive a stabilized image stream, wherein the stabilized image stream is drift-corrected based on determining that an input image stream is stable and then applying drift correction to maintain a stabilized field of view, wherein the field of view is stabilized with respect to the real world. The computing system can operate a vehicle based on determining at least one moving object in the stabilized image stream.

Abstract: A computing system can crop an image based on a width, height and location of a first vehicle in the image. The computing system can estimate a pose of the first vehicle based on inputting the cropped image and the width, height and location of the first vehicle into a deep neural network. The computing system can then operate a second vehicle based on the estimated pose.

Abstract: Techniques and examples pertaining to objection detection and trajectory prediction for autonomous vehicles are described. A processor receives an input stream of image frames and fuses a spatiotemporal input stream of the image frames and an appearance-based stream of the image frames using a deep neural network (DNN) to generate an augmented stream of the image frames. The processor performs object detection and trajectory prediction of one or more objects in the image frames based on the augmented stream.

Abstract: This disclosure generally relates to a system, apparatus, and method for achieving an adaptive vehicle state-based hands free noise reduction feature. A noise reduction tool is provided for adaptively applying a noise reduction strategy on a sound input that uses feedback speech quality measures and machine learning to develop future noise reduction strategies, where the noise reduction strategies include analyzing vehicle operational state information and external information that are predicted to contribute to cabin noise and selecting noise reducing pre-filter options based on the analysis.

Abstract: A processor of a vehicle speech recognition system recognizes speech via domain-specific language and acoustic models. The processor further, in response to the acoustic model having a confidence score for recognized speech falling within a predetermined range defined relative to a confidence score for the domain-specific language model, recognizes speech via the acoustic model only.