Abstract: Provided are a system and methods for a unified framework and tooling for lane boundary annotation, which include obtaining sensor data along a trajectory corresponding to locations of a base map. Features are extracted from the sensor data. The features are input into a trained neural network that outputs overlapping rich feature maps comprising polylines. The overlapping rich feature maps are aggregated according to an aggregation function to obtain raster image. Vectorization is applied to the raster images to extract roadway geometry represented by globally consistent polylines.

Abstract: Provided are methods, systems, and computer program products for generating an output map indicating a likelihood of individual elements of an image as corresponding to particular road elements, such as lane dividers, road dividers, and road boundaries. An example method may include applying a machine learning architecture to the image, which architecture includes a convolutional neural network and a sub-network capturing global context from feature maps generated by the convolutional neural network.

Abstract: A method for determining a trajectory of a vehicle within a physical space based at least on an aligned point cloud may include generating a fused feature map by concatenating a first feature map corresponding to a source point cloud and a second feature map corresponding to a target point cloud. A machine learning model may be applied to determine, based at least on the fused feature map, a relative transform aligning the target point cloud to the source point cloud. An aligned target point cloud may be generated by transforming the target point cloud in accordance with the relative transform. Furthermore, a trajectory of a vehicle within the physical space may be determined based on at least the first relative transform. Related systems and computer program products are also provided.

Abstract: A method for determining a trajectory of a vehicle within a physical space based at least on an aligned point cloud may include generating a fused feature map by concatenating a first feature map corresponding to a source point cloud and a second feature map corresponding to a target point cloud. A machine learning model may be applied to determine, based at least on the fused feature map, a relative transform aligning the target point cloud to the source point cloud. An aligned target point cloud may be generated by transforming the target point cloud in accordance with the relative transform. Furthermore, a trajectory of a vehicle within the physical space may be determined based on at least the first relative transform. Related systems and computer program products are also provided.

Abstract: Provided are methods for enhanced semantic labeling in mapping with a semantic labeling system, which can include receiving, from a LiDAR sensor of a vehicle, LiDAR point cloud information including at least one raw point feature for a point, receiving, from a camera of the vehicle, image data associated with an image captured using the camera, generating at least one rich point feature for the point based on the image data, predicting, using a LiDAR segmentation neural network and based on the at least one raw point feature and the at least one rich point feature, a point-level semantic label for the point, and providing the point-level semantic label to a mapping engine to generate a map based on the point-level semantic label Systems and computer program products are also provided.

Abstract: Provided are techniques for automatic annotation of drivable road segments, including but not limited to: receiving sensor data, generating a map, dividing the map into disjoint sub-maps, inferring sub-map segmentation masks, constructing an inferred segmentation mask, filtering the inferred segmentation mask, smoothing the filtered segmentation mask, planning a path for a vehicle and controller the vehicle.

Abstract: Provided are methods, systems, and computer program products for machine-learning based point cloud alignment classification. An example method may include: obtaining at least two light detection and ranging (LiDAR) point clouds; processing the at least two LiDAR point clouds using at least one classifier network; obtaining at least one output dataset from the at least one classifier network; determining that the at least two LiDAR point clouds are misaligned based on the at least one output dataset; and performing a first action based on the determining that the at least two LiDAR point clouds are misaligned.

Abstract: Provided are methods for semantic annotation of sensor data using unreliable map annotation inputs, which can include training a machine learning model to accept inputs including images representing sensor data for a geographic area and unreliable semantic annotations for the geographic area. The machine learning model can be trained against validated semantic annotations for the geographic area, such that subsequent to training, additional images representing sensor data and additional unreliable semantic annotations can be passed through the neural network to provide predicted semantic annotations for the additional images. Systems and computer program products are also provided.

Abstract: Enclosed are embodiments of an ML-based framework for drivable surface annotation. In an embodiment, a method comprises: obtaining, using at least one processor, multimodal map data for a geographic region; and automatically annotating, using the at least one processor, one or more semantic masks of the map data using a machine learning model.

Abstract: Provided are methods for enhanced semantic labeling in mapping with a semantic labeling system, which can include receiving, from a LiDAR sensor of a vehicle, LiDAR point cloud information including at least one raw point feature for a point, receiving, from a camera of the vehicle, image data associated with an image captured using the camera, generating at least one rich point feature for the point based on the image data, predicting, using a LiDAR segmentation neural network and based on the at least one raw point feature and the at least one rich point feature, a point-level semantic label for the point, and providing the point-level semantic label to a mapping engine to generate a map based on the point-level semantic label Systems and computer program products are also provided.

Abstract: Enclosed are embodiments of an ML-based framework for drivable surface annotation. In an embodiment, a method comprises: obtaining, using at least one processor, multimodal map data for a geographic region; and automatically annotating, using the at least one processor, one or more semantic masks of the map data using a machine learning model.