Age & Gender Recognition
MIT License
python -m pip install -r requirements.txt
It has been tested on these machines :
Run the bash script. This will download and extract the dataset.
bash download.sh
Run train.py
python train.py --input data/imdb.mat
The trained models are stored in the directory checkpoints
as weights.{epoch}-{val_loss}.hdf5
for each epoch if the validation improves over time.
usage: train.py [-h] --input INPUT [--batch_size BATCH_SIZE]
[--nb_epochs NB_EPOCHS] [--validation_split VALIDATION_SPLIT]
python plot_history.py --input models/history.h5
tensordboard --logdir="./logs"
After training the model for 70 epochs, the following results were obtained.
The IMDB-WIKI dataset being used is subject to the following conditions.
Please notice that this dataset is made available for academic research purpose only. All the images are collected from the Internet, and the copyright belongs to the original owners. If any of the images belongs to you and you would like it removed, please kindly inform us, we will remove it from our dataset immediately.
[1] R. Rothe, R. Timofte, and L. V. Gool, "DEX: Deep EXpectation of apparent age from a single image," ICCV, 2015.
[2] R. Rothe, R. Timofte, and L. V. Gool, "Deep expectation of real and apparent age from a single image without facial landmarks," IJCV, 2016.
git checkout -b my-new-feature
git commit -am 'Add some feature'
git push origin my-new-feature
Each project may have many problems. Contributing to the better development of this project by reporting them