This is for demo submission for Singapore Airlines AppChallenge 2018. Team AKOV
Problem statement: Offloaded Passenger Compensation Process.
The analyzer attempts to address customer dissatisfaction at check-in to measure success (for service recovery) upon offloading. This supports the main objective of removing the negative relations that passengers have with being ”bumped off”.

This application utilizes Tensorflow by Google to analyze facial features by customers to deliver insight data as a KPI.
Classifier is trained with a training set of 30,000 facial expression images.

The demo runs on python and utilizes the following dependencies.
Make sure your tensorflow, opencv-python packages are installed.

If not, run the commands :
pip install tensorflow
pip install opencv-python

Run the command: python cs_analyzer.py

Press Esc to escape the app

python retrain.py --output_graph=retrained_graph.pb --output_labels=labels.txt --architecture=MobileNet_1.0_224 --image_dir=images