Human race classification using deep neural networks.
UTKFace dataset is a large-scale face dataset with long age span (range from 0 to 116 years old). The dataset consists of over 20,000 face images with annotations of age, gender, and ethnicity (race).
Preprocessed UTKFace dataset for human race classification task is available on Google.Drive.
The model was trained and tested with Caffe. You can also download and use pre-builded Caffe with CUDA 9.0 and cuDNN 7.6.4 for Windows.
CNN | Solver | Mode | Iterations | Data | Loss-1/Top-1 | Loss-2/Top-1 | Loss-3/Top-1 |
---|---|---|---|---|---|---|---|
GoogleNet | SGD | GPU | 40k | 20k | 0.925 | 0.930 | 0.9325 |
Model snapshot is included.
MIT