Abstract: A device may receive, from a first vehicle, video data for video captured of a location associated with an accident for a second vehicle, and may process the video data, with a first model, to generate a sparse point cloud of the location associated with the accident. The device may process the video data, with a second model, to generate depth maps for frames of the video data, and may utilize the depth maps with the sparse point cloud to generate a dense point cloud. The device may process the video data, with a third model, to generate a dense semantic point cloud, and may process the dense semantic point cloud, with a fourth model, to determine a dense semantic overhead view of the location associated with the accident. The device may perform actions based on the dense semantic overhead view.

Abstract: A scanning laser projector includes an optical module and projection engine. The optical module includes a laser generator outputting a laser beam, and a movable mirror scanning the laser beam across an exit window defined through the housing in a scanning pattern wider than the exit window such that the laser beam is directed through the exit window in a projection pattern that is smaller than and within the scanning pattern. A first light detector is positioned about a periphery of the exit window such that as the movable mirror scans the laser beam in the scan pattern, at a point in the scan pattern where the laser beam is scanned across an interior of the housing and not through the exit window, the laser beam impinges upon the first light detector. The projection engine adjusts driving of the movable mirror based upon output from the first light detector.

Abstract: A scanning laser projector includes an optical module and projection engine. The optical module includes a laser generator outputting a laser beam, and a movable mirror scanning the laser beam across an exit window defined through the housing in a scanning pattern wider than the exit window such that the laser beam is directed through the exit window in a projection pattern that is smaller than and within the scanning pattern. A first light detector is positioned about a periphery of the exit window such that as the movable mirror scans the laser beam in the scan pattern, at a point in the scan pattern where the laser beam is scanned across an interior of the housing and not through the exit window, the laser beam impinges upon the first light detector. The projection engine adjusts driving of the movable mirror based upon output from the first light detector.

Abstract: A scanning laser projector includes an optical module with a housing defined by a top surface, a bottom surface, and sidewalls extending between the top surface and bottom surface to define an interior compartment within the housing. A given one of the sidewalls has an exit window defined therein. A first light detector is positioned at an interior surface of the given one of the sidewalls about a periphery of the exit window. A second light detector positioned at the interior surface of the given one of the sidewalls about the periphery of the exit window and on a different side thereof than the first light detector.

Abstract: A device may receive, from a first vehicle, video data for video captured of a location associated with an accident for a second vehicle, and may process the video data, with a first model, to generate a sparse point cloud of the location associated with the accident. The device may process the video data, with a second model, to generate depth maps for frames of the video data, and may utilize the depth maps with the sparse point cloud to generate a dense point cloud. The device may process the video data, with a third model, to generate a dense semantic point cloud, and may process the dense semantic point cloud, with a fourth model, to determine a dense semantic overhead view of the location associated with the accident. The device may perform actions based on the dense semantic overhead view.

Abstract: A device may receive, from a first vehicle, video data for video captured of a location associated with an accident for a second vehicle, and may process the video data, with a first model, to generate a sparse point cloud of the location associated with the accident. The device may process the video data, with a second model, to generate depth maps for frames of the video data, and may utilize the depth maps with the sparse point cloud to generate a dense point cloud. The device may process the video data, with a third model, to generate a dense semantic point cloud, and may process the dense semantic point cloud, with a fourth model, to determine a dense semantic overhead view of the location associated with the accident. The device may perform actions based on the dense semantic overhead view.

Abstract: A device may receive, from a first vehicle, video data for video captured of a location associated with an accident for a second vehicle, and may process the video data, with a first model, to generate a sparse point cloud of the location associated with the accident. The device may process the video data, with a second model, to generate depth maps for frames of the video data, and may utilize the depth maps with the sparse point cloud to generate a dense point cloud. The device may process the video data, with a third model, to generate a dense semantic point cloud, and may process the dense semantic point cloud, with a fourth model, to determine a dense semantic overhead view of the location associated with the accident. The device may perform actions based on the dense semantic overhead view.