A PyTorch implementation of the DAN, pre-trained models are available for deployment.

• Download pre-trained model of MSCeleb and move the file to ./models
• Download RAF-DB dataset and extract the raf-basic dir to ./datasets
• Download AffectNet dataset and extract the AffectNet dir to ./datasets
• Run python ./utils/convert_affectnet.py to store a split version of AffectNet dataset.

We provide the training code for AffectNet and RAF-DB.

For AffectNet-8 dataset, run:

CUDA_VISIBLE_DEVICES=0 python affectnet.py --epochs 10 --num_class 8

For AffectNet-7 dataset, run:

CUDA_VISIBLE_DEVICES=0 python affectnet.py --epochs 10 --num_class 7

For RAF-DB dataset, run:

CUDA_VISIBLE_DEVICES=0 python rafdb.py

Pre-trained models can be downloaded for evaluation as following:

There is a simple demo to invoke DAN model for a emotion inference:

CUDA_VISIBLE_DEVICES=0 python demo.py --image test_image_path