Abstract: One embodiment of a method for accelerating a trained machine learning model includes parsing the trained machine learning model to identify one or more layers of the trained machine learning model and, for each layer included in the one or more layers, one or more corresponding compression techniques that can be applied to compress the layer, performing, based on a hardware device on which the trained machine learning model is intended to execute, one or more iterative operations to select, for each layer included in the one or more layers, a compression technique and values of one or more parameters associated with the compression technique, and compressing each layer included in the one or more layers using the compression technique that is selected for the layer and the values of the one or more parameters associated with the compression technique to generate a compressed trained machine learning model.

Abstract: Disclosed herein are systems, methods, and computer program products for predicting movement of an object in a real-world environment. The methods comprise: obtaining a plurality of image frames captured in a sequence during a period of time; identifying first image frames of the plurality of image frames that contain an image of at least one object with one or more turn signals; analyzing the first image frames to obtain a classification for a pose of the at least one object; using the classification of the pose of the at least one object to further obtain a type classification for at least one of the turn signals and a state classification for a state of at least one of the turn signals; and predicting movement of the at least one object based at least on the type and state classifications obtained for at least one of the turn signals.

Abstract: Disclosed herein are systems, methods, and computer program products for predicting movement of an object in a real-world environment. The methods comprise: obtaining a plurality of image frames captured in a sequence during a period of time; identifying first image frames of the plurality of image frames that contain an image of at least one object with one or more turn signals; analyzing the first image frames to obtain a classification for a pose of the at least one object; using the classification of the pose of the at least one object to further obtain a type classification for at least one of the turn signals and a state classification for a state of at least one of the turn signals; and predicting movement of the at least one object based at least on the type and state classifications obtained for at least one of the turn signals.

Abstract: A system uses video of a vehicle to detect and classify the vehicle's pose. The system generates an image stack by scaling and shifting a set of digital image frames from the video to a fixed scale, yielding a sequence of images over a time period. The system processes the image stack with a classifier to determine the pose of the object. The system also may determine state and class of visible turn signals on the object, as well as predict the vehicle's direction of travel.

Abstract: A system uses video of a vehicle to detect and classify the vehicle's pose. The system generates an image stack by scaling and shifting a set of digital image frames from the video to a fixed scale, yielding a sequence of images over a time period. The system processes the image stack with a classifier to determine the pose of the object. The system also may determine state and class of visible turn signals on the object, as well as predict the vehicle's direction of travel.

Abstract: A system uses video of a vehicle or other object to detect and classify an active turn sign on the object. The system generates an image stack by scaling and shifting a set of digital image frames from the video to a fixed scale, yielding a sequence of images over a time period. The system processes the image stack with a classifier to determine a pose of the object, as well as the state and class of each visible turn signals on the object. When the system determines that a turn signal is active, the system will predict an action that the object will take based on the class of that signal.

Abstract: A system uses video of a vehicle or other object to detect and classify an active turn sign on the object. The system generates an image stack by scaling and shifting a set of digital image frames from the video to a fixed scale, yielding a sequence of images over a time period. The system processes the image stack with a classifier to determine a pose of the object, as well as the state and class of each visible turn signals on the object. When the system determines that a turn signal is active, the system will predict an action that the object will take based on the class of that signal.