Abstract: Disclosed herein are system, method, and computer program product embodiments for extracting and tracking metadata from a data store. For example, the method includes extracting a plurality of identifiers of data from a data source. An identifier uniquely identifies a record in the data source. The method further includes scanning the data to extract a plurality of data samples, extracting metadata from each data sample of the plurality of data samples, hashing the metadata of each respective data sample to generate a respective hash value associated with each respective data sample of the plurality of data samples, comparing the hash values to identify one or more unique hash values, identifying one or more unique schemas corresponding to the unique hash value, and storing the one or more unique schemas in a data store. The metadata comprises schema indicative of one or more attributes of each respective data sample.

Abstract: In various examples, a neural network may be trained for use in vehicle re-identification tasks—e.g., matching appearances and classifications of vehicles across frames—in a camera network. The neural network may be trained to learn an embedding space such that embeddings corresponding to vehicles of the same identify are projected closer to one another within the embedding space, as compared to vehicles representing different identities. To accurately and efficiently learn the embedding space, the neural network may be trained using a contrastive loss function or a triplet loss function. In addition, to further improve accuracy and efficiency, a sampling technique—referred to herein as batch sample—may be used to identify embeddings, during training, that are most meaningful for updating parameters of the neural network.

Abstract: In various examples, sensor data—such as masked sensor data—may be used as input to a machine learning model to determine a confidence for object to person associations. The masked sensor data may focus the machine learning model on particular regions of the image that correspond to persons, objects, or some combination thereof. In some embodiments, coordinates corresponding to persons, objects, or combinations thereof, in addition to area ratios between various regions of the image corresponding to the persons, objects, or combinations thereof, may be used to further aid the machine learning model in focusing on important regions of the image for determining the object to person associations.

Abstract: A technique, and its applications, for high resolution, rapid, and simple nanopatterning. The general method has been demonstrated in several forms and applications. One is patterning nanophotonic structures at an optical fiber tip for refractive index sensing. Another is patterning nanoresonator structures on a sensor substrate for plasmonic effect related detection of VOCs. In the latter example, a graphene oxide coated plasmonic crystal as a gas sensor capable of identifying different gas species using an array of such structures. By coating the surface of multiple identical plasmonic crystals with different thicknesses of Graphene-Oxide (GO) layer, the effective refractive index of the GO layer on each plasmonic crystal is differently modulated when exposed to a specific gas. Identification of various gas species is accomplished using pattern recognition algorithm.

Abstract: In various examples, a neural network may be trained for use in vehicle re-identification tasks—e.g., matching appearances and classifications of vehicles across frames—in a camera network. The neural network may be trained to learn an embedding space such that embeddings corresponding to vehicles of the same identify are projected closer to one another within the embedding space, as compared to vehicles representing different identities. To accurately and efficiently learn the embedding space, the neural network may be trained using a contrastive loss function or a triplet loss function. In addition, to further improve accuracy and efficiency, a sampling technique—referred to herein as batch sample—may be used to identify embeddings, during training, that are most meaningful for updating parameters of the neural network.

Abstract: Systems may include at least one processor configured to determine a predicted value of an unwrap factor using a machine learning model, wherein the machine learning model is a trained machine learning model configured to provide a predicted value of an unwrap factor for dealiasing a measurement of range rate of a target object as an output, dealiase a measurement value of range rate from a radar of an autonomous vehicle (AV) based on the predicted value of the unwrap factor to provide a true value of range rate, and control an operation of the AV in a real-time environment based on the true value of range rate. Methods, computer program products, and autonomous vehicles are also disclosed.

Abstract: In various examples, sensor data—such as masked sensor data—may be used as input to a machine learning model to determine a confidence for object to person associations. The masked sensor data may focus the machine learning model on particular regions of the image that correspond to persons, objects, or some combination thereof. In some embodiments, coordinates corresponding to persons, objects, or combinations thereof, in addition to area ratios between various regions of the image corresponding to the persons, objects, or combinations thereof, may be used to further aid the machine learning model in focusing on important regions of the image for determining the object to person associations.

Abstract: An example storage medium stores instructions that, when executed, cause a processor of a computing device to receive an indication associated with a first virtual machine, the first virtual machine containing a first application, the indication indicating that a first operation in the first virtual machine is to use a second application; receive information associated with a second virtual machine, the second virtual machine created in response to the first operation and containing the second application; store information describing a chain of virtual machines, the chain of virtual machines including the first and second virtual machines, the stored information including a relationship between the first virtual machine and the second virtual machine, based on the received indication and the received information; and in response to an identification of malware in the chain of virtual machines, identify a particular virtual machine in the chain of virtual machines that is in a kill chain of the malware based

Abstract: Methods of determining relevance of objects that a vehicle detected are disclosed. A system will receive a data log of a run of the vehicle. The data log includes perception data captured by vehicle sensors during the run. The system will identify an interaction time, along with a look-ahead lane based on a lane in which the vehicle traveled during the run. The system will define a region of interest (ROI) that includes a lane segment within the look-ahead lane. The system will identify, from the perception data, objects that the vehicle detected within the ROI during the run. For each object, the system will determine a detectability value by measuring an amount of the object that the vehicle detected. The system will create a subset with only objects having at least a threshold detectability value, and it will classify any such object as a priority relevant object.

Abstract: The present disclosure provides various approaches for smart area monitoring suitable for parking garages or other areas. These approaches may include ROI-based occupancy detection to determine whether particular parking spots are occupied by leveraging image data from image sensors, such as cameras. These approaches may also include multi-sensor object tracking using multiple sensors that are distributed across an area that leverage both image data and spatial information regarding the area, to provide precise object tracking across the sensors. Further approaches relate to various architectures and configurations for smart area monitoring systems, as well as visualization and processing techniques. For example, as opposed to presenting video of an area captured by cameras, 3D renderings may be generated and played from metadata extracted from sensors around the area.

Abstract: An example storage medium includes instructions that, when executed, cause a processor of a computing device to encrypt a source file that has been identified as potentially malicious, place the encrypted file in a location accessible to a virtual machine, provide, to the virtual machine, information for decrypting the encrypted file, and cause the virtual machine to use the information to process the encrypted file.

Abstract: The present disclosure provides various approaches for smart area monitoring suitable for parking garages or other areas. These approaches may include ROI-based occupancy detection to determine whether particular parking spots are occupied by leveraging image data from image sensors, such as cameras. These approaches may also include multi-sensor object tracking using multiple sensors that are distributed across an area that leverage both image data and spatial information regarding the area, to provide precise object tracking across the sensors. Further approaches relate to various architectures and configurations for smart area monitoring systems, as well as visualization and processing techniques. For example, as opposed to presenting video of an area captured by cameras, 3D renderings may be generated and played from metadata extracted from sensors around the area.

Abstract: An example non-transitory computer readable storage medium comprises instructions that when executed cause a processor of an electronic device to: in response to detecting a malware scan trigger associated with a file, determine a combined risk score associated with the file based on metadata of the file and a source of the malware scan trigger, where the source includes a file access interceptor, a file write observer, and a file indexer; determine a scan priority based on the combined risk score; and perform a malware scan on the file based on the scan priority.

Abstract: A method provides power from battery units by placing units with a predetermined variance in voltages in a first configuration; detecting a divergence in module voltages; if the divergence crosses a threshold, creating a new configuration of units to provide an even voltage distribution; and electrically rerouting the units to form the new configuration while the battery units are in an idle state or in a reduced mode of operation.

Abstract: In various examples, sensor data—such as masked sensor data—may be used as input to a machine learning model to determine a confidence for object to person associations. The masked sensor data may focus the machine learning model on particular regions of the image that correspond to persons, objects, or some combination thereof. In some embodiments, coordinates corresponding to persons, objects, or combinations thereof, in addition to area ratios between various regions of the image corresponding to the persons, objects, or combinations thereof, may be used to further aid the machine learning model in focusing on important regions of the image for determining the object to person associations.

Abstract: Systems and methods are disclosed for controlling packs of rechargeable batteries by: for each pack: determining electrical characteristics of the pack; determining energy correction for the pack; calculating a voltage correction for the pack; and determining a dispatch modifier for the pack. The system then normalizes the dispatch modifiers for all packs and recalculates the dispatch modifiers for all packs.

Abstract: In various examples, sensor data—such as masked sensor data—may be used as input to a machine learning model to determine a confidence for object to person associations. The masked sensor data may focus the machine learning model on particular regions of the image that correspond to persons, objects, or some combination thereof. In some embodiments, coordinates corresponding to persons, objects, or combinations thereof, in addition to area ratios between various regions of the image corresponding to the persons, objects, or combinations thereof, may be used to further aid the machine learning model in focusing on important regions of the image for determining the object to person associations.

Abstract: Systems may include at least one processor configured to determine a predicted value of an unwrap factor using a machine learning model, wherein the machine learning model is a trained machine learning model configured to provide a predicted value of an unwrap factor for dealiasing a measurement of range rate of a target object as an output, dealiase a measurement value of range rate from a radar of an autonomous vehicle (AV) based on the predicted value of the unwrap factor to provide a true value of range rate, and control an operation of the AV in a real-time environment based on the true value of range rate. Methods, computer program products, and autonomous vehicles are also disclosed.

Abstract: Disclosed herein are systems, methods, and computer program products for operating a robotic system. For example, the method includes: obtaining a first cuboid generated based on an image, a second cuboid generated based on a lidar dataset and/or a third cuboid generated by a heuristic algorithm using the lidar dataset; using a machine learning model to generate a heading for an object in proximity to the robotic system based on the first cuboid, second cuboid and/or third cuboid; generating a bounding box geometry and a bounding box location based on the second cuboid or third cuboid; and generating a fourth cuboid using the bounding box geometry, the bounding box location, and the heading generated using the machine learning model.

Abstract: A computing device comprises a memory to store a first untrusted file and a second untrusted file; and a processor to scan a file system operation executing on the computing device; create an association between the first untrusted file and the second untrusted file based on the scanning; execute the first untrusted file together with the associated second untrusted file in a micro virtual machine (VM); and identify a malicious behavior of the executed first untrusted file interacting with the associated second untrusted file in the micro VM.