Abstract: Systems, devices and processes and described herein to improve the generation of video content in video production systems. In general, the systems, devices and processes facilitate the incorporation of one or more remotely captured video streams into an output video stream for use in video conferencing. Specifically, the systems, devices and processes provide for the use of a dynamically adjustable time clock delay to generate a relatively low latency output video stream suitable for use in video conferencing. The use of dynamically adjustable time clock delay allows one or more input video streams to be generated by one or more remote video-capture devices, transmitted to a video access point device, and selectively combined at the video access point device into the output video stream with relatively low latency suitable for video conferencing.

Abstract: Systems, devices and processes and described herein to improve the generation of video content in video production systems. In general, the systems, devices and processes facilitate the incorporation of one or more remotely captured video streams into an output video stream for use in video conferencing. Specifically, the systems, devices and processes provide for the use of a dynamically adjustable time clock delay to generate a relatively low latency output video stream suitable for use in video conferencing. The use of dynamically adjustable time clock delay allows one or more input video streams to be generated by one or more remote video capture devices, transmitted to a video access point device, and selectively combined at the video access point device into the output video stream with relatively low latency suitable for video conferencing.

Abstract: This disclosure relates to a system and method for attention-based surface crack segmentation. Existing methods do not efficiently handle the sub-problem of data imbalance and inaccurate predicted pixels are ignored. The present disclosure obtains a binary edge map by passing a m-channel image through an edge detection algorithm and concatenate the obtained binary edge map with a channel dimension to obtain a (m+1)-channel image. Feature maps are extracted from an encoder and a decoder by feeding the obtained (m+1)-channel image into a network, wherein the feature maps are convolved with an attention mask and merged in a fused network. The merged feature maps are up sampled and concatenated to obtain a final fused feature map. The final fused feature map is passed through a sigmoid activation function to obtain a probability map which is iteratively thresholded to obtain a binary predicted image. The binary image is indicative of crack pixels.

Abstract: This disclosure relates generally to multi instance visual tracking based on observer motion modelling. Conventional state-of-the-art methods have major challenges in dealing with the error rate, accuracy of tracking the moving objects when observer motion is also considered. Embodiments of the present disclosure provide a method for multi instance visual tracking considering the motion of observer. The method tracks moving objects in an input video comprising a set of image frames, by detecting the moving objects in each image frame. The method detects the moving objects in each image frame by calculating trifocal tensors and fundamental matrix of the previous frames. The disclosed multi instance visual tracking can be used in remote surveillance problem for tracking moving objects.

Abstract: Systems, devices and processes and described herein to improve the generation of video content in video production systems. In general, the systems, devices and processes facilitate the incorporation of one or more remotely captured video streams into an output video stream for use in video conferencing. Specifically, the systems, devices and processes provide for the use of a dynamically adjustable time clock delay to generate a relatively low latency output video stream suitable for use in video conferencing. The use of dynamically adjustable time clock delay allows one or more input video streams to be generated by one or more remote video capture devices, transmitted to a video access point device, and selectively combined at the video access point device into the output video stream with relatively low latency suitable for video conferencing.

Abstract: This disclosure relates generally to multi instance visual tracking based on observer motion modelling. Conventional state-of-the-art methods have major challenges in dealing with the error rate, accuracy of tracking the moving objects when observer motion is also considered. Embodiments of the present disclosure provide a method for multi instance visual tracking considering the motion of observer. The method tracks moving objects in an input video comprising a set of image frames, by detecting the moving objects in each image frame. The method detects the moving objects in each image frame by calculating trifocal tensors and fundamental matrix of the previous frames. The disclosed multi instance visual tracking can be used in remote surveillance problem for tracking moving objects.

Abstract: The systems and methods of the present disclosure provide a path panning algorithm for fixed-wing aerial vehicles that may be employed, particularly for monitoring of long linear infrastructures. The applicants' earlier patent applications address turn angle constraints for fixed wing aerial and maintaining transmission continuity in presence of coverage holes by imposing a plurality of constraints along with storage constraints. The present disclosure addresses a technical challenge of simultaneously meeting multiple objectives; particularly distance cost and communication cost while satisfying the plurality of constraints that enable pruning of feasible paths in a 3D Euclidean navigation space to obtain a set of optimal paths for surveillance of a target under consideration.

Abstract: Remote sensing techniques are being increasingly used for periodic structural health monitoring of vast infrastructures. Conventionally, analysis of visual and other signals captured from sensing devices are used to diagnose faults. Such data collection and analysis is expensive in terms of both computational overheads as well as towards robotic maneuvering of data collection systems, such as a UAV. In accordance with the present disclosure, the data acquisition system is modeled as an intelligent situated agent that autonomously controls data gathering and analysis activities through a cognitive cycle of perception-recognition-action, in order to optimize the cost of efforts in identifying faults that may exist. Also, a reactive, economical planning algorithm around Qualitative Bayesian Network (QBN) that controls the sequence of data collection and analysis has been implemented.

Abstract: System(s) and method(s) for real-time data communication over an Internet of Things (IoT) network are described. According to the present subject matter, the system(s) implement the described method(s) for real-time data communication over the IoT network. The method includes encoding, at a source communication device, data to be exchanged between peer sub-layers of IoT entities based on a Forward Error Correction (FEC) context to generate encoded data packets, the IoT entities comprising the source communication device and a destination communication device. The method further includes identifying time delay to be maintained for transmission of the encoded data packets from the source communication device to the destination communication device to have minimal data packet drop due to queue overflow at the source communication device. The method further includes transmitting the encoded data packets over the IoT network.

Abstract: A computer implemented system for providing robust communication links to unmanned aerial vehicles is envisaged. It comprises a plurality of nodes which communicate with each other and with an unmanned aerial vehicle to allow exchange of data. A 3D signal coverage model is created which determines signal coverage provided by the plurality of nodes. A navigator present in the system navigates the unmanned aerial vehicle to follow a stored flight path based on this 3D model. Waypoints present in the path of the unmanned aerial vehicle are then identified and suitable waypoints are selected from where sensed pre-stored data is collected. A suitable node is then selected based on the stored 3D signal coverage model, location of the unmanned aerial vehicle and the nodes, and the signal strength of the nodes and the collected data is transmitted to the suitable node through the unmanned aerial vehicle to provide robust communication.

Abstract: Remote sensing techniques are being increasingly used for periodic structural health monitoring of vast infrastructures. Conventionally, analysis of visual and other signals captured from sensing devices are used to diagnose faults. Such data collection and analysis is expensive in terms of both computational overheads as well as towards robotic maneuvering of data collection systems, such as a UAV. In accordance with the present disclosure, the data acquisition system is modeled as an intelligent situated agent that autonomously controls data gathering and analysis activities through a cognitive cycle of perception-recognition-action, in order to optimize the cost of efforts in identifying faults that may exist. Also, a reactive, economical planning algorithm around Qualitative Bayesian Network (QBN) that controls the sequence of data collection and analysis has been implemented.

Abstract: The present disclosure provides systems and methods to enable detection of linear objects such as utility poles in complex and heterogeneous outdoor surroundings. The methods deal with shape and orientation as prominent features of a pole model. Candidate trapeziums from 2D images of the face of the poles are extracted, some of which represent parts of the pole. To overcome the missed detection of certain parts due to problems of occlusion and diffusion into background, shape based techniques, that extrapolate and capture a longer trapezium representing the pole is implemented. The region growing stage or extrapolation is driven by orientation-based clustering of trapeziums. Context information is further used to identify objects of interest, by discarding false positives. Besides detecting poles of interest, the detected poles are further analyzed to identify damages, if any.

Abstract: The systems and methods of the present disclosure provide a path panning algorithm for fixed-wing aerial vehicles that may be employed, particularly for monitoring of long linear infrastructures. The applicants' earlier patent applications address turn angle constraints for fixed wing aerial and maintaining transmission continuity in presence of coverage holes by imposing a plurality of constraints along with storage constraints. The present disclosure addresses a technical challenge of simultaneously meeting multiple objectives; particularly distance cost and communication cost while satisfying the plurality of constraints that enable pruning of feasible paths in a 3D Euclidean navigation space to obtain a set of optimal paths for surveillance of a target under consideration.

Abstract: This disclosure relates generally to visual inspection systems, and more particularly to a method and system for facilitating real time detection of linear infrastructural objects in aerial imagery. In one embodiment, a background suppression technique is applied to one or more hardware processors to a HSV image. Further, a mean shift filtering technique is applied to the hardware processors to find a peak of a confidence map and then a gradient image generation is performed for a plurality of edges of the image. A seed point pair along a middle cut portion of a linear feature of the HSV image to identify one or more boundaries of the seed point pair is extracted and then a contour growing approach to detect the boundaries of the linear feature is initiated. Lastly, one or more false positives are removed by using a rigidity feature, the rigidity feature being equivalent to the total sum of gradient orientations.

Abstract: This disclosure relates generally to Unmanned Aerial Vehicle (UAV), and more particularly to system and a method for landing of an Unmanned Aerial Vehicle (UAV). In one embodiment, the method includes estimating a 3-dimensional (3D) location of at least one media sensor mounted on the UAV relative to a marker representative of a landing location of the UAV. The marker comprises a recursive geometrical pattern. The landing of the UAV on the marker at the landing location is facilitated based on the 3D location of the at least one media sensor mounted on the UAV relative to the marker.

Abstract: This disclosure relates generally to ge visual inspection systems, and more particularly to a method and system for facilitating real time detection of linear infrastructural objects in aerial imagery. In one embodiment, a background suppression technique is applied to one or more hardware processors to a HSV image. Further, a mean shift filtering technique is applied to the hardware processors to find a peak of a confidence map and then a gradient image generation is performed for a plurality of edges of the image. A seed point pair along a middle cut portion of a linear feature of the HSV image to identify one or more boundaries of the seed point pair is extracted and then a contour growing approach to detect the boundaries of the linear feature is initiated. Lastly, one or more false positives are removed by using a rigidity feature, the rigidity feature being equivalent to the total sum of gradient orientations.

Abstract: The present disclosure provides systems and methods to enable detection of linear objects such as utility poles in complex and heterogeneous outdoor surroundings. The methods deal with shape and orientation as prominent features of a pole model. Candidate trapeziums from 2D images of the face of the poles are extracted, some of which represent parts of the pole. To overcome the missed detection of certain parts due to problems of occlusion and diffusion into background, shape based techniques, that extrapolate and capture a longer trapezium representing the pole is implemented. The region growing stage or extrapolation is driven by orientation-based clustering of trapeziums. Context information is further used to identify objects of interest, by discarding false positives. Besides detecting poles of interest, the detected poles are further analysed to identify damages, if any.

Abstract: This disclosure relates generally to Unmanned Aerial Vehicle (UAV), and more particularly to system and a method for landing of an Unmanned Aerial Vehicle (UAV). In one embodiment, the method includes estimating a 3-dimensional (3D) location of at least one media sensor mounted on the UAV relative to a marker representative of a landing location of the UAV. The marker comprises a recursive geometrical pattern. The landing of the UAV on the marker at the landing location is facilitated based on the 3D location of the at least one media sensor mounted on the UAV relative to the marker.

Abstract: A computer implemented system for providing robust communication links to unmanned aerial vehicles is envisaged. It comprises a plurality of nodes which communicate with each other and with an unmanned aerial vehicle to allow exchange of data. A 3D signal coverage model is created which determines signal coverage provided by the plurality of nodes. A navigator present in the system navigates the unmanned aerial vehicle to follow a stored flight path based on this 3D model. Waypoints present in the path of the unmanned aerial vehicle are then identified and suitable waypoints are selected from where sensed pre-stored data is collected. A suitable node is then selected based on the stored 3D signal coverage model, location of the unmanned aerial vehicle and the nodes, and the signal strength of the nodes and the collected data is transmitted to the suitable node through the unmanned aerial vehicle to provide robust communication.

Abstract: System(s) and method(s) for real-time data communication over an Internet of Things (IoT) network are described. According to the present subject matter, the system(s) implement the described method(s) for real-time data communication over the IoT network. The method includes encoding, at a source communication device, data to be exchanged between peer sub-layers of IoT entities based on a Forward Error Correction (FEC) context to generate encoded data packets, the IoT entities comprising the source communication device and a destination communication device. The method further includes identifying time delay to be maintained for transmission of the encoded data packets from the source communication device to the destination communication device to have minimal data packet drop due to queue overflow at the source communication device. The method further includes transmitting the encoded data packets over the IoT network.