Abstract: A data processing solution for data generated by autonomous vehicles, unmanned aerial vehicles (UAVs), and other Internet of Things (IoT) devices may include a network of edge devices and one or more cloud servers to gather and process data. Initially, all data generated by the edge devices may be gathered, transported, and processed by the cloud server(s) to train machine learning models. The cloud server(s) may push machine learning (ML) models to the network of edge devices. Using the ML models running on the edge devices, the data gathered by the individual edge devices may be filtered by its relevance score, which correlates to the impact the data would have as training data. The relevance score is used by the edge device to identify the data to upload to the cloud.

Abstract: A system is configured to label computer vision datasets using eye tracking of users that track objects depicted in imagery to label the datasets. The imagery may include moving images (e.g., video) or still images. By using eye tracking, users may be able to label large amounts of imagery more efficiently than by manually labeling datasets using conventional input devices. A user may be instructed to watch a particular object during a playback of the video while an imaging device determines a direction of the user's gaze which correlates with a location in the imagery. An application may then associate the location in the imagery determined from the user's gaze as a location of the object on a frame-by-frame basis, or for certain frames.

Abstract: Described are automated systems and methods for providing a simulated environment for robotic and/or real-time systems, such as unmanned vehicles, to perform full system simulations while also providing a system-wide code coverage assessment of the software associated with the robotic and/or real-time systems. The exemplary systems and methods can employ code coverage instrumented shared libraries to facilitate generation of code coverage information and one or more code coverage reports. The code coverage information and/or the code coverage report can quantify the effectiveness of the testing and can facilitate development of more comprehensive and efficient testing of the software.

Abstract: A system is configured to label computer vision datasets using eye tracking of users that track objects depicted in imagery to label the datasets. The imagery may include moving images (e.g., video) or still images. By using eye tracking, users may be able to label large amounts of imagery more efficiently than by manually labeling datasets using conventional input devices. A user may be instructed to watch a particular object during a playback of the video while an imaging device determines a direction of the user's gaze which correlates with a location in the imagery. An application may then associate the location in the imagery determined from the user's gaze as a location of the object on a frame-by-frame basis, or for certain frames.