Abstract: A method for automatic generation of calibration parameters for a surround view (SV) camera system is provided that includes capturing a video stream from each camera comprised in the SV camera system, wherein each video stream captures two calibration charts in a field of view of the camera generating the video stream; displaying the video streams in a calibration screen on a display device coupled to the SV camera system, wherein a bounding box is overlaid on each calibration chart, detecting feature points of the calibration charts, displaying the video streams in the calibration screen with the bounding box overlaid on each calibration chart and detected features points overlaid on respective calibration charts, computing calibration parameters based on the feature points and platform dependent parameters comprising data regarding size and placement of the calibration charts, and storing the calibration parameters in the SV camera system.

Abstract: Aspects of the subject disclosure relate to a vehicle camera system. A device implementing the subject technology includes a first set of cameras and a processor configured to receive first data from at least one camera of a second set of cameras of an object configured to be towed by a vehicle and second data from at least one camera of the first set of cameras. The processor may determine, using a trained machine learning algorithm, a set of sub-pixel shift values that represent relative positions of images in the first data and the second data based on a position of the at least one camera of the first set of cameras. The processor may align, using the trained machine learning algorithm, the images based on the set of sub-pixel shift values, and combine the aligned images to produce a stitched image having a combined field of view.

Abstract: Aspects of the subject disclosure relate to a vehicle camera system. A device implementing the subject technology may include a processor configured to detect a first camera on a vehicle and a second camera on the vehicle. The processor can determine a position and a pose of at least one of the first camera or the second camera. The processor also can receive first data from the first camera and second data from the second camera. The processor can stitch the first data with the second data to generate a stitched image having a contiguous field of view based on the position and the pose of the at least one of the first camera or the second camera. Accordingly, seamless integration of cameras between a vehicle and towed objects of various geometries for surround view can be provided.

Abstract: Systems and methods for determining the drivable space of a road, for applications such as autonomous navigation. To determine the non-drivable space under another vehicle, systems and methods of embodiments of the disclosure generate 3D bounding boxes from 2D bounding boxes of objects in captured roadway images, and from various geometric constraints. Image portions may be labeled as drivable or non-drivable according to projections of these 3D bounding boxes onto their road surfaces. These labeled images, along with accompanying semantic information, may be compiled to form training datasets for a machine learning model such as a CNN. The training datasets may train the CNN to classify input image portions into drivable and non-drivable space, for applications such as autonomous navigation.

Abstract: A system for determining blockage of a vehicle camera is configured to apply a partition grid having a plurality of locations to a sequence of images to form a plurality of regions. The system determines at least one spatial feature corresponding to the partition grid and at least one temporal feature corresponding to the partition grid. The system generates a sequence of classifications for each location of the plurality of locations based on the at least one spatial feature, the at least one temporal feature, and reference information. The system applies a smoothing technique to determine a subset of regions that are blocked among the sequence of classifications, and generates an output signal based on the subset of regions. The output may be provided to an output system to wash the camera lens, notify the user or the vehicle of the blockage, or modify image processing.

Abstract: A method for automatic generation of calibration parameters for a surround view (SV) camera system is provided that includes capturing a video stream from each camera comprised in the SV camera system, wherein each video stream captures two calibration charts in a field of view of the camera generating the video stream; displaying the video streams in a calibration screen on a display device coupled to the SV camera system, wherein a bounding box is overlaid on each calibration chart, detecting feature points of the calibration charts, displaying the video streams in the calibration screen with the bounding box overlaid on each calibration chart and detected features points overlaid on respective calibration charts, computing calibration parameters based on the feature points and platform dependent parameters comprising data regarding size and placement of the calibration charts, and storing the calibration parameters in the SV camera system.

Abstract: A system on a chip (SoC) includes a digital signal processor (DSP) and a graphics processing unit (GPU) coupled to the DSP. The DSP is configured to receive a stream of received depth measurements and generate a virtual bowl surface based on the stream of received depth measurements. The DSP is also configured to generate a bowl to physical camera mapping based on the virtual bowl surface. The GPU is configured to receive a first texture and receive a second texture. The GPU is also configured to perform physical camera to virtual camera transformation on the first texture and on the second texture, based on the bowl to physical camera mapping, to generate an output image.

Abstract: A method for automatic generation of calibration parameters for a surround view (SV) camera system is provided that includes capturing a video stream from each camera comprised in the SV camera system, wherein each video stream captures two calibration charts in a field of view of the camera generating the video stream; displaying the video streams in a calibration screen on a display device coupled to the SV camera system, wherein a bounding box is overlaid on each calibration chart, detecting feature points of the calibration charts, displaying the video streams in the calibration screen with the bounding box overlaid on each calibration chart and detected features points overlaid on respective calibration charts, computing calibration parameters based on the feature points and platform dependent parameters comprising data regarding size and placement of the calibration charts, and storing the calibration parameters in the SV camera system.

Abstract: Systems and methods for performing auto-calibration for a target sensor from information gained by a reference sensor, thereby enabling cross-sensor self-calibration for a set of sensors. A method includes the step of obtaining a first mapping of a first observable scene from a first scene sensing device and a second mapping of a second observable scene from a second scene sensing device. The first and second scene sensing devices are positioned on a test object (e.g., vehicle) in a substantially fixed relationship with respect to each other. The first and second observable scenes have common reference objects and/or overlapping portions. The method also includes comparing the common reference object and/or overlapping portion of the first mapping with the common reference object and/or overlapping portion of the second mapping to determine a difference. Also, the method includes creating calibration parameters, based on the difference, for calibrating the second scene sensing device.

Abstract: Apparatuses, systems, and methods for determining a field in a path of travel of a vehicle for detection of objects in the path of travel. An apparatus includes a path identifier configured to identify a path of travel for a vehicle. At least one camera is disposed on the vehicle and is configured to capture image data of an area including the path of travel. A region of interest selector is configured to select a region of interest within the image data. A horizon identifier is configured to identify a horizon in the image data. A field determiner is configured to project the horizon onto the region of interest to isolate a field specified by the path of travel and the horizon, the field being evaluatable for the presence of objects in the path of travel of the vehicle.

Abstract: Disclosed embodiments include apparatuses, systems, and methods for determining a field in a path of travel of a vehicle for detection of objects in the path of travel. In an illustrative embodiment, an apparatus includes a path identifier configured to identify a path of travel for a vehicle. At least one camera is disposed on the vehicle and is configured to capture image data of an area including the path of travel. A region of interest selector is configured to select a region of interest within the image data. A horizon identifier is configured to identify a horizon in the image data. A field determiner is configured to project the horizon onto the region of interest to isolate a field specified by the path of travel and the horizon, the field being evaluatable for the presence of objects in the path of travel of the vehicle.

Abstract: Systems and methods for determining the drivable space of a road, for applications such as autonomous navigation. To determine the non-drivable space under another vehicle, systems and methods of embodiments of the disclosure generate 3D bounding boxes from 2D bounding boxes of objects in captured roadway images, and from various geometric constraints. Image portions may be labeled as drivable or non-drivable according to projections of these 3D bounding boxes onto their road surfaces. These labeled images, along with accompanying semantic information, may be compiled to form training datasets for a machine learning model such as a CNN. The training datasets may train the CNN to classify input image portions into drivable and non-drivable space, for applications such as autonomous navigation.

Abstract: A system on a chip (SoC) includes a digital signal processor (DSP) and a graphics processing unit (GPU) coupled to the DSP. The DSP is configured to receive a stream of received depth measurements and generate a virtual bowl surface based on the stream of received depth measurements. The DSP is also configured to generate a bowl to physical camera mapping based on the virtual bowl surface. The GPU is configured to receive a first texture and receive a second texture. The GPU is also configured to perform physical camera to virtual camera transformation on the first texture and on the second texture, based on the bowl to physical camera mapping, to generate an output image.

Abstract: A system for determining blockage of a vehicle camera is configured to apply a partition grid having a plurality of locations to a sequence of images to form a plurality of regions. The system determines at least one spatial feature corresponding to the partition grid and at least one temporal feature corresponding to the partition grid. The system generates a sequence of classifications for each location of the plurality of locations based on the at least one spatial feature, the at least one temporal feature, and reference information. The system applies a smoothing technique to determine a subset of regions that are blocked among the sequence of classifications, and generates an output signal based on the subset of regions. The output may be provided to an output system to wash the camera lens, notify the user or the vehicle of the blockage, or modify image processing.

Abstract: A system on a chip (SoC) includes a digital signal processor (DSP) and a graphics processing unit (GPU) coupled to the DSP. The DSP is configured to receive a stream of received depth measurements and generate a virtual bowl surface based on the stream of received depth measurements. The DSP is also configured to generate a bowl to physical camera mapping based on the virtual bowl surface. The GPU is configured to receive a first texture and receive a second texture. The GPU is also configured to perform physical camera to virtual camera transformation on the first texture and on the second texture, based on the bowl to physical camera mapping, to generate an output image.

Abstract: Disclosed embodiments include apparatuses, systems, and methods for determining a field in a path of travel of a vehicle for detection of objects in the path of travel. In an illustrative embodiment, an apparatus includes a path identifier configured to identify a path of travel for a vehicle. At least one camera is disposed on the vehicle and is configured to capture image data of an area including the path of travel. A region of interest selector is configured to select a region of interest within the image data. A horizon identifier is configured to identify a horizon in the image data. A field determiner is configured to project the horizon onto the region of interest to isolate a field specified by the path of travel and the horizon, the field being evaluatable for the presence of objects in the path of travel of the vehicle.

Abstract: A method for automatic generation of calibration parameters for a surround view (SV) camera system is provided that includes capturing a video stream from each camera comprised in the SV camera system, wherein each video stream captures two calibration charts in a field of view of the camera generating the video stream; displaying the video streams in a calibration screen on a display device coupled to the SV camera system, wherein a bounding box is overlaid on each calibration chart, detecting feature points of the calibration charts, displaying the video streams in the calibration screen with the bounding box overlaid on each calibration chart and detected features points overlaid on respective calibration charts, computing calibration parameters based on the feature points and platform dependent parameters comprising data regarding size and placement of the calibration charts, and storing the calibration parameters in the SV camera system.

Abstract: A method for automatic generation of calibration parameters for a surround view (SV) camera system is provided that includes capturing a video stream from each camera comprised in the SV camera system, wherein each video stream captures two calibration charts in a field of view of the camera generating the video stream; displaying the video streams in a calibration screen on a display device coupled to the SV camera system, wherein a bounding box is overlaid on each calibration chart, detecting feature points of the calibration charts, displaying the video streams in the calibration screen with the bounding box overlaid on each calibration chart and detected features points overlaid on respective calibration charts, computing calibration parameters based on the feature points and platform dependent parameters comprising data regarding size and placement of the calibration charts, and storing the calibration parameters in the SV camera system.

Abstract: A system on a chip (SoC) includes a digital signal processor (DSP) and a graphics processing unit (GPU) coupled to the DSP. The DSP is configured to receive a stream of received depth measurements and generate a virtual bowl surface based on the stream of received depth measurements. The DSP is also configured to generate a bowl to physical camera mapping based on the virtual bowl surface. The GPU is configured to receive a first texture and receive a second texture. The GPU is also configured to perform physical camera to virtual camera transformation on the first texture and on the second texture, based on the bowl to physical camera mapping, to generate an output image.

Abstract: In some embodiments, a computer-readable medium stores executable code, which, when executed by a processor, causes the processor to: capture an image of a finder pattern using a camera; locate a predetermined point within the finder pattern; use the predetermined point to identify multiple boundary points on a perimeter associated with the finder pattern; identify fitted boundary lines based on the multiple boundary points; and locate feature points using the fitted boundary lines.