Abstract: Provided are methods for end-to-end perception system training using fused images, which can include fusing different types of images to form a fused image, extracting features from the fused image, calculating a loss, and modifying at least one network parameter of an image semantic network based on the loss. Systems and computer program products are also provided.

Abstract: Techniques for operation of a vehicle using machine learning with motion planning include storing, using one or more processors of a vehicle located within an environment, a plurality of constraints for operating the vehicle within the environment. One or more sensors of the vehicle receive sensor data describing the environment. The one or more processors extract a feature vector from the stored plurality of constraints and the received sensor data. The feature vector includes a first feature describing an object located within the environment. A machine learning circuit of the vehicle is used to generate a first motion segment based on the feature vector. A number of violations of the stored plurality of constraints is below a threshold. The one or more processors operate the vehicle in accordance with the generated first motion segment.

Abstract: A system receives a 3D image having multiple data points, and uses one or more filters, such as a distance filter, map filter, and/or height filter to remove certain 3D data points from the image. The system may group the data points and annotate them to identify unknown or unclassified objects within the image.

Abstract: Provided are methods for end-to-end perception system training using fused images, which can include fusing different types of images to form a fused image, extracting features from the fused image, calculating a loss, and modifying at least one network parameter of an image semantic network based on the loss. Systems and computer program products are also provided.

Abstract: Techniques for operation of a vehicle using machine learning with motion planning include storing, using one or more processors of a vehicle located within an environment, a plurality of constraints for operating the vehicle within the environment. One or more sensors of the vehicle receive sensor data describing the environment. The one or more processors extract a feature vector from the stored plurality of constraints and the received sensor data. The feature vector includes a first feature describing an object located within the environment. A machine learning circuit of the vehicle is used to generate a first motion segment based on the feature vector. A number of violations of the stored plurality of constraints is below a threshold. The one or more processors operate the vehicle in accordance with the generated first motion segment.

Abstract: Techniques for operation of a vehicle using machine learning with motion planning include storing, using one or more processors of a vehicle located within an environment, a plurality of constraints for operating the vehicle within the environment. One or more sensors of the vehicle receive sensor data describing the environment. The one or more processors extract a feature vector from the stored plurality of constraints and the received sensor data. The feature vector includes a first feature describing an object located within the environment. A machine learning circuit of the vehicle is used to generate a first motion segment based on the feature vector. A number of violations of the stored plurality of constraints is below a threshold. The one or more processors operate the vehicle in accordance with the generated first motion segment.

Abstract: Provided are methods for motion prediction in an autonomous vehicle using fused synthetic and camera images. The method can include obtaining data pairs, each of which reflects data corresponding to a synthetic image representing a birds-eye-view of an area around a vehicle and identifying an object, and data corresponding to a camera image depicting the object. A machine learning model can be trained based on the data pairs to result in a trained model that predicts motion of the object within the data pair based on the synthetic image and camera image in the data pair. Systems and computer program products are also provided.

Abstract: Disclosed is an apparatus including an input interface. The input interface is configured to obtain input sensor data. For example, the input sensor data includes light detection and ranging (LiDAR) data indicative of an environment. The apparatus includes a pre-processor communicatively coupled to the input interface. The pre-processor optionally includes a parser logic. The pre-processor includes a decoder communicatively coupled to the parser logic. The segmentation logic is configured to parse the input sensor data. The decoder is configured to decode the parsed input sensor data. The pre-processor is configured to provide the decoded input sensor data and the parsed input sensor data to a processor for fusion.

Abstract: Among other things, techniques for detecting objects in the environment surrounding a vehicle are described. A computer system is configured to receive a set of measurements from a sensor of a vehicle. The set of measurements includes a plurality of data points that represent a plurality of objects in a 3D space surrounding the vehicle. The system divides the 3D space into a plurality of pillars. The system then assigns each data point of the plurality of data points to a pillar in the plurality of pillars. The system generates a pseudo-image based on the plurality of pillars. The pseudo-image includes, for each pillar of the plurality of pillars, a corresponding feature representation of data points assigned to the pillar. The system detects the plurality of objects based on an analysis of the pseudo-image. The system then operates the vehicle based upon the detecting of the objects.

Abstract: Provided are methods for customized tags for annotating sensor data, which can include receiving sensor data captured during a plurality of sensor data capture sessions, processing the sensor data using a plurality of machine learning models to identify a plurality of capture session collections represented in the sensor data, filtering the sensor data based at least partly on a user-specified category of the plurality of categories of capture session to identify a capture session collection, of the plurality of capture session collections, representing sensor data of one or more sensor data capture sessions that conforms to the user-specified category, and transmitting the sensor data of one or more sensor data capture sessions that conforms to the user-specified category to an end user computing device. Systems and computer program products are also provided.

Abstract: A system is configured to segment instances of objects in a pseudo-image. The pseudo-image can be generated from a 3D image coupled to a vehicle. The system can receive the pseudo-image, which can include multiple sections. The system can determine an object classification for a section and determine an instance portion classification for the section. The system can group the section with another section based on the object classification and the instance portion classification. The grouping can correspond to an instance of an object in the image. The system can use the grouping to navigate the vehicle.

Abstract: Techniques are provided for improving a perception processing pipeline for object detection that fuses image segmentation data (e.g., segmentation scores) with LiDAR points. The disclosed techniques are implemented using an architecture that accepts point clouds and images as input and estimates oriented 3D bounding boxes for all relevant object classes. In an embodiment, a method comprises: matching temporally, using one or more processors of a vehicle, points in a three-dimensional (3D) point cloud with an image; generating, using an image-based neural network, semantic data for the image; decorating, using the one or more processors, the points in the 3D point cloud with the semantic data; and estimating, using a 3D object detector with the decorated points as input, oriented 3D bounding boxes for the one or more objects.

Abstract: Among other things, techniques for detecting objects in the environment surrounding a vehicle are described. A computer system is configured to receive a set of measurements from a sensor of a vehicle. The set of measurements includes a plurality of data points that represent a plurality of objects in a 3D space surrounding the vehicle. The system divides the 3D space into a plurality of pillars. The system then assigns each data point of the plurality of data points to a pillar in the plurality of pillars. The system generates a pseudo-image based on the plurality of pillars. The pseudo-image includes, for each pillar of the plurality of pillars, a corresponding feature representation of data points assigned to the pillar. The system detects the plurality of objects based on an analysis of the pseudo-image. The system then operates the vehicle based upon the detecting of the objects.

Abstract: Among other things, techniques for detecting objects in the environment surrounding a vehicle are described. A computer system is configured to receive a set of measurements from a sensor of a vehicle. The set of measurements includes a plurality of data points that represent a plurality of objects in a 3D space surrounding the vehicle. The system divides the 3D space into a plurality of pillars. The system then assigns each data point of the plurality of data points to a pillar in the plurality of pillars. The system generates a pseudo-image based on the plurality of pillars. The pseudo-image includes, for each pillar of the plurality of pillars, a corresponding feature representation of data points assigned to the pillar. The system detects the plurality of objects based on an analysis of the pseudo-image. The system then operates the vehicle based upon the detecting of the objects.

Abstract: Provided are methods for end-to-end perception system training using fused images, which can include fusing different types of images to form a fused image, extracting features from the fused image, calculating a loss, and modifying at least one network parameter of an image semantic network based on the loss. Systems and computer program products are also provided.

Abstract: Provided are methods for customized tags for annotating sensor data, which can include receiving sensor data captured during a plurality of sensor data capture sessions, processing the sensor data using a plurality of machine learning models to identify a plurality of capture session collections represented in the sensor data, filtering the sensor data based at least partly on a user-specified category of the plurality of categories of capture session to identify a capture session collection, of the plurality of capture session collections, representing sensor data of one or more sensor data capture sessions that conforms to the user-specified category, and transmitting the sensor data of one or more sensor data capture sessions that conforms to the user-specified category to an end user computing device. Systems and computer program products are also provided.

Abstract: Techniques for operation of a vehicle using machine learning with motion planning include storing, using one or more processors of a vehicle located within an environment, a plurality of constraints for operating the vehicle within the environment. One or more sensors of the vehicle receive sensor data describing the environment. The one or more processors extract a feature vector from the stored plurality of constraints and the received sensor data. The feature vector includes a first feature describing an object located within the environment. A machine learning circuit of the vehicle is used to generate a first motion segment based on the feature vector. A number of violations of the stored plurality of constraints is below a threshold. The one or more processors operate the vehicle in accordance with the generated first motion segment.

Abstract: Techniques for operation of a vehicle using machine learning with motion planning include storing, using one or more processors of a vehicle located within an environment, a plurality of constraints for operating the vehicle within the environment. One or more sensors of the vehicle receive sensor data describing the environment. The one or more processors extract a feature vector from the stored plurality of constraints and the received sensor data. The feature vector includes a first feature describing an object located within the environment. A machine learning circuit of the vehicle is used to generate a first motion segment based on the feature vector. A number of violations of the stored plurality of constraints is below a threshold. The one or more processors operate the vehicle in accordance with the generated first motion segment.

Abstract: Techniques are provided for improving a perception processing pipeline for object detection that fuses image segmentation data (e.g., segmentation scores) with LiDAR points. The disclosed techniques are implemented using an architecture that accepts point clouds and images as input and estimates oriented 3D bounding boxes for all relevant object classes. In an embodiment, a method comprises: matching temporally, using one or more processors of a vehicle, points in a three-dimensional (3D) point cloud with an image; generating, using an image-based neural network, semantic data for the image; decorating, using the one or more processors, the points in the 3D point cloud with the semantic data; and estimating, using a 3D object detector with the decorated points as input, oriented 3D bounding boxes for the one or more objects.

Abstract: Techniques are provided for improving a perception processing pipeline for object detection that fuses image segmentation data (e.g., segmentation scores) with LiDAR points. The disclosed techniques are implemented using an architecture that accepts point clouds and images as input and estimates oriented 3D bounding boxes for all relevant object classes. In an embodiment, a method comprises: matching temporally, using one or more processors of a vehicle, points in a three-dimensional (3D) point cloud with an image; generating, using an image-based neural network, semantic data for the image; decorating, using the one or more processors, the points in the 3D point cloud with the semantic data; and estimating, using a 3D object detector with the decorated points as input, oriented 3D bounding boxes for the one or more objects.