Chonnam National University
2017 AI Class
Professor Lee Chilwoo
Student: Nguyen Hai Duong
Target: predict human gender using Convolutional Neural Network
- Download the Jupyter notebook
gender_prediction.ipynb
and save it to a specific path (called GD_PATH) - In GD_PATH, create folder gender/wiki_crop
- Download the processed WIKI dataset [1] and save them to GD_PATH/gender/wiki_crop
- Download additional testing images and store them in GD_PATH
- Open
gender_prediction.ipynb
with Jupyter and run all cells
- Download the Jupyter notebook
gender_prediction_testing.ipynb
and save it to a specific path (called GD_PATH) - Download trained model on WIKI dataset [1] and save it to GD_PATH
- In GD_PATH, create folder gender/wiki_crop
- Download WIKI testingset [1] including 64_64_11938_4098_testing_x_onehot.npy, and 64_64_11938_4098_testing_y_onehot.npy and store them in GD_PATH/gender/wiki_crop
- Download additional testing images and store them in GD_PATH
- Open
gender_prediction_testing.ipynb
with Jupyter and run all cells
[1] Rasmus Rothe, Radu Timofte, and Luc Van Gool, "Deep expectation of real and apparent age from a single image without facial landmarks," International Journal of Computer Vision (2016)
Nguyen Hai Duong
Supervisor: Professor Kim Soo-Hyung
Pattern Recognition Lab
Chonnam National University, Korea
E-mail: nhduong_3010@live.com