Abstract: A method for detecting drowsy driving in a vehicle is provided. The method includes receiving and processing images from a camera within the vehicle. The processed images result in a sequence of facial images, which are stored in a memory buffer. A selection of images from these facial images constitutes a video. The processor then determines a drowsy confidence score based on the video. If the drowsy confidence score crosses a predetermined threshold, an alert is generated.

Abstract: A method for detecting a driving event includes receiving video data from one or more vehicle-mounted cameras. The method includes generating one or more video embeddings in an embedding space based on the video data. The method includes generating one or more text embeddings of a natural language input in the embedding space. The natural language input corresponds to a driving event description. The method includes comparing the one or more video embeddings and the one or more text embeddings in the embedding space. The method includes determining at least one similarity value between the one or more video embeddings and the one or more text embeddings based on the comparison. The method includes detecting the driving event when the at least one similarity value is above a detection threshold.

Abstract: A computing device (120) can monitor an in-vehicle temperature of a vehicle (110). The computing device can receive one or ore images (302, 304, 306) of an environment within the vehicle (110) depicting a driver of the vehicle. The computing device can determine the in-vehicle temperature of the vehicle satisfies a temperature threshold. The computing device can execute a machine learning model (140) using at least the one or more images to detect the driver is in a thermal distress state. The computing device can generate a message in the vehicle requesting driver input. The computing device can generate a real-time video feed (301) depicting the driver to a remote computing system. By limiting the image processing to instances in which the in-vehicle temperature satisfies the temperature threshold, the computing device can increase the accuracy of the machine learning model and reduce the frequency of thermal-distress related alerts.

Abstract: Systems and methods are provided for detecting and classifying driving behaviors, as well as techniques for improving such systems and methods, which may include identifying geographic locations where anomalous rates of non-compliant driving behaviors are observed, such as in the vicinity of certain traffic signs, traffic lights, and road markings.

Abstract: Systems and methods are provided for intelligent driving monitoring systems, advanced driver assistance systems and autonomous driving systems, and providing alerts to the driver of a vehicle. Combinations of co-occurring driving events may be detected and used to warn on anomalies, prevent accidents, provide feedback to the driver, and in general provide a safer driver experience.

Abstract: Systems and methods are provided for intelligent driving monitoring systems, advanced driver assistance systems and autonomous driving systems, and providing alerts to the driver of a vehicle, based on anomalies detected between driver behavior and environment captured by the outward facing camera. Various aspects of the driver, which may include his direction of sight, point of focus, posture, gaze, is determined by image processing of the upper visible body of the driver, by a driver facing camera in the vehicle. Other aspects of environment around the vehicle captured by the multitude of cameras in the vehicle are used to correlate driver behavior and actions with what is happening outside to detect and warn on anomalies, prevent accidents, provide feedback to the driver, and in general provide a safer driver experience.

Abstract: Systems and methods are provided for intelligent driving monitoring systems, advanced driver assistance systems and autonomous driving systems, and providing alerts to the driver of a vehicle. Combinations of co-occurring driving events may be detected and used to warn on anomalies, prevent accidents, provide feedback to the driver, and in general provide a safer driver experience.

Abstract: A system and method for deploying machine learning models with consistent fixed-point arithmetic processing. A computing device in a vehicle receives a trained machine learning model that was trained using training images processed with fixed-point arithmetic operations. The computing device receives images from a camera and processes the received images using fixed-point arithmetic operations that are consistent with those used during training. The machine learning model is executed using the processed images to detect objects such as vehicles, pedestrians, persons on bikes, stop lights, or stop signs. The computing device may comprise specialized hardware including a graphics processing unit (GPU), digital signal processing hardware decoder, or single-purpose hardware decoder such as an application-specific integrated circuit (ASIC) or field-programmable gate array (FPGA).

Abstract: Systems, devices and methods provide, implement, and use vision-based methods of sequence inference for a device affixed to a vehicle.

Abstract: A system or method trains and implements a machine learning model for object detection. A first computing device remote to a vehicle trains a machine learning model by receiving a decoded set of previously encoded training image data and training the machine learning model to detect objects within images based on the decoded set of training image data. A processor of a second computing device encodes at least one image captured by a camera, decodes the encoded image, and executes the machine learning model to extract a set of objects from the decoded image.

Abstract: A system or method trains and implements a machine learning model for object detection. A first computing device remote to a vehicle trains a machine learning model by receiving a decoded set of previously encoded training image data and training the machine learning model to detect objects within images based on the decoded set of training image data. A processor of a second computing device encodes at least one image captured by a camera, decodes the encoded image, and executes the machine learning model to extract a set of objects from the decoded image.

Abstract: Disclosed herein are systems and methods for classifying traffic indicators to detect intersection crossings by a vehicle. A computing device can receive a sequence of frames captured by a capture device mounted to the vehicle, and generate an intersection status data structure for each frame in the sequence of frames using a full-frame classification model. The computing device can also classify a set of features detected in each frame using an object detection model, and detect an intersection crossing event based on the intersection status data structure of each frame and the classification of each of the set of features detected in each frame. The systems and methods described herein improve upon existing traffic light classification models by using a full-frame intersection classification model in connection with traditional object tracking techniques, thereby reducing false-positive crossing-event detection and improving the accuracy of classification in a variety of environmental conditions.

Abstract: Systems and methods are provided for intelligent driving monitoring systems, advanced driver assistance systems and autonomous driving systems, and providing alerts to the driver of a vehicle, based on anomalies detected between driver behavior and environment captured by the outward facing camera. Various aspects of the driver, which may include his direction of sight, point of focus, posture, gaze, is determined by image processing of the upper visible body of the driver, by a driver facing camera in the vehicle. Other aspects of environment around the vehicle captured by the multitude of cameras in the vehicle are used to correlate driver behavior and actions with what is happening outside to detect and warn on anomalies, prevent accidents, provide feedback to the driver, and in general provide a safer driver experience.

Abstract: Disclosed herein are systems and methods for classifying traffic indicators to detect intersection crossings by a vehicle. A computing device can receive a sequence of frames captured by a capture device mounted to the vehicle, and generate an intersection status data structure for each frame in the sequence of frames using a full-frame classification model. The computing device can also classify a set of features detected in each frame using an object detection model, and detect an intersection crossing event based on the intersection status data structure of each frame and the classification of each of the set of features detected in each frame. The systems and methods described herein improve upon existing traffic light classification models by using a full-frame intersection classification model in connection with traditional object tracking techniques, thereby reducing false-positive crossing-event detection and improving the accuracy of classification in a variety of environmental conditions.

Abstract: Systems and methods are provided for classifying one or more images of video data captured by a camera in a vehicle. The classification involves a detection of a traffic control device, such as a U-Turn sign, in image data. The classification further involves querying a database with a location of the vehicle from which the traffic control device was observed. In some embodiments, the database contains entries for various types of traffic control devices from a family of traffic control devices, where each type of traffic control device in the family has a similar visual appearance. The database query may therefore help disambiguate the visually perceived traffic control device at a particular location. Embodiments may be incorporated into navigation or driver safety systems.

Abstract: Disclosed herein are systems and methods for classifying traffic indicators to detect intersection crossings by a vehicle. A computing device can receive a sequence of frames captured by a capture device mounted to the vehicle, and generate an intersection status data structure for each frame in the sequence of frames using a full-frame classification model. The computing device can also classify a set of features detected in each frame using an object detection model, and detect an intersection crossing event based on the intersection status data structure of each frame and the classification of each of the set of features detected in each frame. The systems and methods described herein improve upon existing traffic light classification models by using a full-frame intersection classification model in connection with traditional object tracking techniques, thereby reducing false-positive crossing-event detection and improving the accuracy of classification in a variety of environmental conditions.

Abstract: Systems and methods are provided for detecting and classifying driving behaviors, as well as techniques for improving such systems and methods, which may include identifying geographic locations where anomalous rates of non-compliant driving behaviors are observed, such as in the vicinity of certain traffic signs, traffic lights, and road markings.

Abstract: Systems and methods are provided for intelligent driving monitoring systems, advanced driver assistance systems and autonomous driving systems, and providing alerts to the driver of a vehicle, based on anomalies detected between driver behavior and environment captured by the outward facing camera. Various aspects of the driver, which may include his direction of sight, point of focus, posture, gaze, is determined by image processing of the upper visible body of the driver, by a driver facing camera in the vehicle. Other aspects of environment around the vehicle captured by the multitude of cameras in the vehicle are used to correlate driver behavior and actions with what is happening outside to detect and warn on anomalies, prevent accidents, provide feedback to the driver, and in general provide a safer driver experience.

Abstract: Systems, devices and methods provide, implement, and use vision-based methods of sequence inference for a device affixed to a vehicle.

Abstract: The present disclosure provides devices, systems and methods for mapping of traffic boundaries.