Age and Gender Recognition Software

  • Implemented a neural network architecture based on deep CNN in Caffe to predict the age and gender of a face image;

  • Improved the accuracy of age and gener prediction from 50.7%, 86.8% to 79.6%, 95.2% respectively;

  • Developed a software on Windows and graphic user interface by C#, caught the face image appearing in a photo or video, and introduce advertisements based on predicted age and gender;

  • Open Source