Thank you, Prof Teo Yong Meng and TA Sunimal, Zhang Han

An emoji says a thousand words. Emojis play a role in text messaging just like the tone of a phone call or the gesture during communication. We are going to show the result of facial expression recognition in a more intuitive and interesting way.

• Image acquisition - Upload an image from user or take a photo in real time.
• Face recognition - Use API of face recognition from IBM Watson.
• Emotion presentation - Read the person’s expression and show a related Emoji.
• Gender identification - Recognize the person’s gender.
• Age prediction - Estimate an accurate range of the person’s age.

• Obtain pictures - use camera to take a photo or upload a picture.
• Detect face &expression - use the models from IBM Cloud services which are trained by dataset of KDEF with almost two thousand photos.
• Match - match a corresponding emoji according to the detected expression.

• Cloud Platform - IBM Bluemix
• Built as a SaaS application
• Languages Used
  • Back-end – Node.js
  • Front-end – HTML, CSS, JavaScript
• Datasets Used
  • Face recognition dataset – KDEF
  • Emoji dataset contains different emojis

• Doctors can judge patients' physical state according to their detected facial expression and conduct corresponding medical treatment in real time.
• Shopping malls can capture images of customers through cameras and analyze their facial expressions, and further interpret customers' emotional information to analyze customers' satisfaction in shopping malls according to users' mood.

• Have learned more about cloud computing, including its concepts, models, technologies and its application with big data.
• Have learned how to implement the interaction between front-end and back-end with Node.js.
• Have learned how to use the service from IBM cloud and push our web-based big data application to cloud.
• Have learned how to work as a team.