Excited to share a demo of our senior design project developed for the U.S. Navy, a real-time drone detection and tracking system built using computer vision and simulation. The system demonstrates multi-camera detection across varying conditions including fog, rain, snow, day and night environments. Recorded on a GTX 1660 Super, optimized for real-time performance on higher-end hardware. Source code proprietary. Open to discussing the project further! #ComputerVision #MachineLearning #ObjectDetection #SeniorDesign #Engineering #DeepLearning #YOLO #UnrealEngine #PyTorch #DroneDetection

Great demo, and amazing optimization on only a GTX 1660 super

Like
Reply

Next is running a quantified comparison of realism between real-world environments and the simulator!

Like
Reply

Really well done good job!

Like
Reply

Nathan Dessi A multi-camera system is not all-weather, has a big data processing challenge, and as a result, is slow. Absence of reliable data is increasing AI delay and the reliability of results. A staring antenna array can cover and simultaneously track multiple objects as minimum of three orders fates and more reliably. System providing high sensitivity and can be invisible, passive. Connect if interested.

  • No alternative text description for this image
Like
Reply
See more comments

To view or add a comment, sign in

Explore content categories