Abstract: The present invention extends to methods, systems, and computer program products for classifying time series image data. Aspects of the invention include encoding motion information from video frames in an eccentricity map. An eccentricity map is essentially a static image that aggregates apparent motion of objects, surfaces, and edges, from a plurality of video frames. In general, eccentricity reflects how different a data point is from the past readings of the same set of variables. Neural networks can be trained to detect and classify actions in videos from eccentricity maps. Eccentricity maps can be provided to a neural network as input. Output from the neural network can indicate if detected motion in a video is or is not classified as an action, such as, for example, a hand gesture.

Abstract: The present invention extends to methods, systems, and computer program products for classifying time series image data. Aspects of the invention include encoding motion information from video frames in an eccentricity map. An eccentricity map is essentially a static image that aggregates apparent motion of objects, surfaces, and edges, from a plurality of video frames. In general, eccentricity reflects how different a data point is from the past readings of the same set of variables. Neural networks can be trained to detect and classify actions in videos from eccentricity maps. Eccentricity maps can be provided to a neural network as input. Output from the neural network can indicate if detected motion in a video is or is not classified as an action, such as, for example, a hand gesture.

Abstract: The present invention extends to methods, systems, and computer program products for classifying time series image data. Aspects of the invention include encoding motion information from video frames in an eccentricity map. An eccentricity map is essentially a static image that aggregates apparent motion of objects, surfaces, and edges, from a plurality of video frames. In general, eccentricity reflects how different a data point is from the past readings of the same set of variables. Neural networks can be trained to detect and classify actions in videos from eccentricity maps. Eccentricity maps can be provided to a neural network as input. Output from the neural network can indicate if detected motion in a video is or is not classified as an action, such as, for example, a hand gesture.

Abstract: A vehicle controller receives sensor outputs and identifies features in the sensor outputs. The controller determines a trajectory based on the features and generates control signals to vehicle actuators to follow the trajectory. Eccentricity of the control signals is evaluated and if it meets a threshold condition is met an intervention is performed such as discarding or modifying the control signal or initiating a safety procedure. Eccentricity may be determined using an unsupervised machine learning model. The threshold condition may be a dynamic threshold condition such as using the n-sigma approach or the Chebyshev inequality.

Abstract: A vehicle controller receives sensor outputs and identifies features in the sensor outputs. The controller determines a trajectory based on the features and generates control signals to vehicle actuators to follow the trajectory. Eccentricity of the control signals is evaluated and if it meets a threshold condition is met an intervention is performed such as discarding or modifying the control signal or initiating a safety procedure. Eccentricity may be determined using an unsupervised machine learning model. The threshold condition may be a dynamic threshold condition such as using the n-sigma approach or the Chebyshev inequality.

Abstract: The present invention extends to methods, systems, and computer program products for classifying time series image data. Aspects of the invention include encoding motion information from video frames in an eccentricity map. An eccentricity map is essentially a static image that aggregates apparent motion of objects, surfaces, and edges, from a plurality of video frames. In general, eccentricity reflects how different a data point is from the past readings of the same set of variables. Neural networks can be trained to detect and classify actions in videos from eccentricity maps. Eccentricity maps can be provided to a neural network as input. Output from the neural network can indicate if detected motion in a video is or is not classified as an action, such as, for example, a hand gesture.

Abstract: A vehicle controller receives sensor outputs and identifies features in the sensor outputs. The controller determines a trajectory based on the features and generates control signals to vehicle actuators to follow the trajectory. Eccentricity of the control signals is evaluated and if it meets a threshold condition is met an intervention is performed such as discarding or modifying the control signal or initiating a safety procedure. Eccentricity may be determined using an unsupervised machine learning model. The threshold condition may be a dynamic threshold condition such as using the n-sigma approach or the Chebyshev inequality.