Intelligent lost luggage tracking system
SITA is logy Company with over $1 billion in revenue, building advanced solutions for airports)
When airline passengers’ luggage is lost, the goal is to find lost luggage at airports using video feeds.
Searching terabytes of videos manually is a very laborious task.
The supplier created a deep learning model to find missing baggage, with OCR (optical character recognition) to extract information from it.
For background segmentation, they used FgSegNet and for finding baggage we used Triplet Network and Siamese Network.
They achieved a Top-5 accuracy of 94%.
The final solution was built in C++ with a user interface.
The solution works real time with use of Jetson GPU’s.