Abstract: Example embodiments of the present disclosure provide an example computer-implemented method for constructing a three-dimensional semantic segmentation of a scene from two-dimensional inputs. The example method includes obtaining, by a computing system comprising one or more processors, an image set comprising one or more views of a subject scene. The example method includes generating, by the computing system and based at least in part on the image set, a scene representation describing the subject scene in three dimensions. The example method includes generating, by the computing system and using a machine-learned semantic segmentation model framework, a multidimensional field of probability distributions over semantic categories, the multidimensional field defined over the three dimensions of the subject scene. The example method includes outputting, by the computing system, classification data for at least one location in the subject scene.

Abstract: Provided are machine learning models that generate geometry-free neural scene representations through efficient object-centric novel-view synthesis. In particular, one example aspect of the present disclosure provides a novel framework in which an encoder model (e.g., an encoder transformer network) processes one or more RGB images (with or without pose) to produce a fully latent scene representation that can be passed to a decoder model (e.g., a decoder transformer network). Given one or more target poses, the decoder model can synthesize images in a single forward pass. In some example implementations, because transformers are used rather than convolutional or MLP networks, the encoder can learn an attention model that extracts enough 3D information about a scene from a small set of images to render novel views with correct projections, parallax, occlusions, and even semantics, without explicit geometry.

Abstract: Systems, methods, and apparatus for identifying and tracking UAVs including an image capturing device. A computing device can receive a frame captured via an image capturing device configured to monitor a particular air space. The computing device can identify a region of interest (ROI) in the frame. The ROI can include an image of an object. The computing device can perform a background subtraction process on the frame. The computing device can scale the frame to a uniform size. The computing device can perform a comparison of the frame to reference images. The reference images can include known unmanned aerial vehicle (UAV) images and known non-UAV images. The computing device can classify the object with a UAV classification based on the comparison.

Abstract: Systems, methods, and apparatus for identifying and tracking UAVs including a computing device and a Wi-Fi sensor. The computing device can receive Wi-Fi data from the Wi-Fi sensor comprising an RSSI and a MAC address. The computing device can determine an estimated proximity of an unmanned aerial vehicle (UAV) based on the RSSI. The computing device can compare the estimated proximity to a signal threshold. The computing device can determine whether the MAC address matches one of a plurality of known UAV MAC addresses. The computing device can apply rule set to determine an action to take. The computing device can perform the action.

Abstract: Systems, methods, and apparatus for identifying and tracking UAVs including a plurality of sensors operatively connected over a network to a configuration of software and/or hardware. A computing device can tune the RF receiver to a particular frequency set. The computing device can receive RF signal data corresponding to a plurality of RF signals via the RF receiver. The computing device can detect a plurality of signal characteristics corresponding to the plurality of RF signals from the RF signal data. The computing device can identify a matching RF signal by comparing the RF signal data to a plurality of known RF signals. The computing device can apply a predetermined rule set to the matching RF signal to determine at least one action to take.

Abstract: Systems and methods for detecting radio frequency (“RF”) signals and corresponding origination locations are disclosed. An RF sensor device includes a software-defined radio and an antenna pair for receiving RF signals. Furthermore the RF sensor device may include a processing unit for processing/analyzing the RF signals, or the processing unit may be remote. The system calculates a phase difference between an RF signal received at two separate antennas of an antenna pair. The phase difference, the distance between the antennas, and the frequency of the RF signal are used for determining the origination direction of the RF signal. In various embodiments, the origination direction may indicate the location of a UAV controller or base station. The software-defined radio may include more than one antenna pair, connected to multiplexers, for efficiently scanning different frequencies by alternating active antenna pairs. Moreover, the system may execute packet-based processing on the RF signal data.

Abstract: Systems, methods, and apparatus for performing an action when an aggregated confidence measure. Data is received from a first sensor proximate to a particular air space. Data is also received from a second sensor and a third sensor proximate to the particular air space. The data from the first sensor, second sensor, and third sensor are each analyzed to determine respective confidence measures that a UAV is within the particular air space. The first sensor corresponds to a first type of data, the second sensor corresponds to a second type of data, and the third sensor corresponds to a third type of data. The confidence measures from each sensor are aggregated together to generate a combined confidence measure indicating a possible presence of the UAV within in the particular air space. When the combined confidence measure exceeds a threshold, an action is taken.

Abstract: Systems and methods for detecting radio frequency (“RF”) signals and corresponding origination locations are disclosed. An RF sensor device includes a software-defined radio and an antenna pair for receiving RF signals. Furthermore the RF sensor device may include a processing unit for processing/analyzing the RF signals, or the processing unit may be remote. The system calculates a phase difference between an RF signal received at two separate antennas of an antenna pair. The phase difference, the distance between the antennas, and the frequency of the RF signal are used for determining the origination direction of the RF signal. In various embodiments, the origination direction may indicate the location of a UAV controller or base station. The software-defined radio may include more than one antenna pair, connected to multiplexers, for efficiently scanning different frequencies by alternating active antenna pairs. Moreover, the system may execute packet-based processing on the RF signal data.

Abstract: Example embodiments of the present disclosure provide an example computer-implemented method for constructing a three-dimensional semantic segmentation of a scene from two-dimensional inputs. The example method includes obtaining, by a computing system comprising one or more processors, an image set comprising one or more views of a subject scene. The example method includes generating, by the computing system and based at least in part on the image set, a scene representation describing the subject scene in three dimensions. The example method includes generating, by the computing system and using a machine-learned semantic segmentation model framework, a multidimensional field of probability distributions over semantic categories, the multidimensional field defined over the three dimensions of the subject scene. The example method includes outputting, by the computing system, classification data for at least one location in the subject scene.

Abstract: Provided are machine learning models that generate geometry-free neural scene representations through efficient object-centric novel-view synthesis. In particular, one example aspect of the present disclosure provides a novel framework in which an encoder model (e.g., an encoder transformer network) processes one or more RGB images (with or without pose) to produce a fully latent scene representation that can be passed to a decoder model (e.g., a decoder transformer network). Given one or more target poses, the decoder model can synthesize images in a single forward pass. In some example implementations, because transformers are used rather than convolutional or MLP networks, the encoder can learn an attention model that extracts enough 3D information about a scene from a small set of images to render novel views with correct projections, parallax, occlusions, and even semantics, without explicit geometry.

Abstract: Systems and methods for detecting radio frequency (“RF”) signals and corresponding origination locations are disclosed. An RF sensor device includes a software-defined radio and an antenna pair for receiving RF signals. Furthermore the RF sensor device may include a processing unit for processing/analyzing the RF signals, or the processing unit may be remote. The system calculates a phase difference between an RF signal received at two separate antennas of an antenna pair. The phase difference, the distance between the antennas, and the frequency of the RF signal are used for determining the origination direction of the RF signal. In various embodiments, the origination direction may indicate the location of a UAV controller or base station. The software-defined radio may include more than one antenna pair, connected to multiplexers, for efficiently scanning different frequencies by alternating active antenna pairs. Moreover, the system may execute packet-based processing on the RF signal data.

Abstract: Systems and methods for detecting radio frequency (“RF”) signals and corresponding origination locations are disclosed. An RF sensor device includes a software-defined radio and an antenna pair for receiving RF signals. Furthermore the RF sensor device may include a processing unit for processing/analyzing the RF signals, or the processing unit may be remote. The system calculates a phase difference between an RF signal received at two separate antennas of an antenna pair. The phase difference, the distance between the antennas, and the frequency of the RF signal are used for determining the origination direction of the RF signal. In various embodiments, the origination direction may indicate the location of a UAV controller or base station. The software-defined radio may include more than one antenna pair, connected to multiplexers, for efficiently scanning different frequencies by alternating active antenna pairs. Moreover, the system may execute packet-based processing on the RF signal data.

Abstract: Systems, methods, and apparatus for performing an action when an aggregated confidence measure. Data is received from a first sensor proximate to a particular air space. Data is also received from a second sensor and a third sensor proximate to the particular air space. The data from the first sensor, second sensor, and third sensor are each analyzed to determine respective confidence measures that a UAV is within the particular air space. The first sensor corresponds to a first type of data, the second sensor corresponds to a second type of data, and the third sensor corresponds to a third type of data. The confidence measures from each sensor are aggregated together to generate a combined confidence measure indicating a possible presence of the UAV within in the particular air space. When the combined confidence measure exceeds a threshold, an action is taken.

Abstract: Systems, methods, and apparatus for identifying and tracking UAVs including a plurality of sensors operatively connected over a network to a configuration of software and/or hardware. A computing device can tune the RF receiver to a particular frequency set. The computing device can receive RF signal data corresponding to a plurality of RF signals via the RF receiver. The computing device can detect a plurality of signal characteristics corresponding to the plurality of RF signals from the RF signal data. The computing device can identify a matching RF signal by comparing the RF signal data to a plurality of known RF signals. The computing device can apply a predetermined rule set to the matching RF signal to determine at least one action to take.

Abstract: Systems, methods, and apparatus for identifying and tracking UAVs including a plurality of sensors operatively connected over a network to a configuration of software and/or hardware. Generally, the plurality of sensors monitors a particular environment and transmits the sensor data to the configuration of software and/or hardware. The data from each individual sensor can be directed towards a process configured to best determine if a UAV is present or approaching the monitored environment. The system generally allows for a detected UAV to be tracked, which may allow for the system or a user of the system to predict how the UAV will continue to behave over time. The sensor information as well as the results generated from the systems and methods may be stored in one or more databases in order to improve the continued identifying and tracking of UAVs.

Abstract: Systems, methods, and apparatus for identifying and tracking UAVs including a plurality of sensors operatively connected over a network to a configuration of software and/or hardware. Generally, the plurality of sensors monitors a particular environment and transmits the sensor data to the configuration of software and/or hardware. The data from each individual sensor can be directed towards a process configured to best determine if a UAV is present or approaching the monitored environment. The system generally allows for a detected UAV to be tracked, which may allow for the system or a user of the system to predict how the UAV will continue to behave over time. The sensor information as well as the results generated from the systems and methods may be stored in one or more databases in order to improve the continued identifying and tracking of UAVs.

Abstract: Systems, methods, and apparatus for identifying and tracking UAVs including a plurality of sensors operatively connected over a network to a configuration of software and/or hardware. Generally, the plurality of sensors monitors a particular environment and transmits the sensor data to the configuration of software and/or hardware. The data from each individual sensor can be directed towards a process configured to best determine if a UAV is present or approaching the monitored environment. The system generally allows for a detected UAV to be tracked, which may allow for the system or a user of the system to predict how the UAV will continue to behave over time. The sensor information as well as the results generated from the systems and methods may be stored in one or more databases in order to improve the continued identifying and tracking of UAVs.

Abstract: Systems, methods, and apparatus for identifying and tracking UAVs including a plurality of sensors operatively connected over a network to a configuration of software and/or hardware. Generally, the plurality of sensors monitors a particular environment and transmits the sensor data to the configuration of software and/or hardware. The data from each individual sensor can be directed towards a process configured to best determine if a UAV is present or approaching the monitored environment. The system generally allows for a detected UAV to be tracked, which may allow for the system or a user of the system to predict how the UAV will continue to behave over time. The sensor information as well as the results generated from the systems and methods may be stored in one or more databases in order to improve the continued identifying and tracking of UAVs.

Abstract: Systems, methods, and apparatus for identifying and tracking UAVs including a plurality of sensors operatively connected over a network to a configuration of software and/or hardware. Generally, the plurality of sensors monitors a particular environment and transmits the sensor data to the configuration of software and/or hardware. The data from each individual sensor can be directed towards a process configured to best determine if a UAV is present or approaching the monitored environment. The system generally allows for a detected UAV to be tracked, which may allow for the system or a user of the system to predict how the UAV will continue to behave over time. The sensor information as well as the results generated from the systems and methods may be stored in one or more databases in order to improve the continued identifying and tracking of UAVs.

Abstract: Systems, methods, and apparatus for identifying and tracking UAVs including a plurality of sensors operatively connected over a network to a configuration of software and/or hardware. Generally, the plurality of sensors monitors a particular environment and transmits the sensor data to the configuration of software and/or hardware. The data from each individual sensor can be directed towards a process configured to best determine if a UAV is present or approaching the monitored environment. The system generally allows for a detected UAV to be tracked, which may allow for the system or a user of the system to predict how the UAV will continue to behave over time. The sensor information as well as the results generated from the systems and methods may be stored in one or more databases in order to improve the continued identifying and tracking of UAVs.