A PyTorch implementation of MPIIGaze and MPIIFaceGaze

Here is a demo program.

Requirements

pip install -r requirements.txt

bash scripts/download_mpiigaze_dataset.sh
python tools/preprocess_mpiigaze.py --dataset datasets/MPIIGaze -o datasets/

bash scripts/download_mpiifacegaze_dataset.sh
python tools/preprocess_mpiifacegaze.py --dataset datasets/MPIIFaceGaze_normalized -o datasets/

This repository uses YACS for configuration management. Default parameters are specified in gaze_estimation/config/defaults.py (which is not supposed to be modified directly). You can overwrite those default parameters using a YAML file like configs/mpiigaze/lenet_train.yaml.

By running the following code, you can train a model using all the data except the person with ID 0, and run test on that person.

python train.py --config configs/mpiigaze/lenet_train.yaml
python evaluate.py --config configs/mpiigaze/lenet_eval.yaml

Using scripts/run_all_mpiigaze_lenet.sh and scripts/run_all_mpiigaze_resnet_preact.sh, you can run all training and evaluation for LeNet and ResNet-8 with default parameters.

Model	Mean Test Angle Error [degree]	Training Time
LeNet	6.52	3.5 s/epoch
ResNet-preact-8	5.73	7 s/epoch

The training time is the value when using GTX 1080Ti.

Model	Mean Test Angle Error [degree]	Training Time
AlexNet	5.06	135 s/epoch
ResNet-14	4.83	62 s/epoch

The training time is the value when using GTX 1080Ti.

This demo program runs gaze estimation on the video from a webcam.

Download the dlib pretrained model for landmark detection.
```
bash scripts/download_dlib_model.sh
```
Calibrate the camera.

Save the calibration result in the same format as the sample file data/calib/sample_params.yaml.
Run demo.

Specify the model path and the path of the camera calibration results in the configuration file as in configs/demo_mpiigaze_resnet.yaml.
```
python demo.py --config configs/demo_mpiigaze_resnet.yaml
```

Zhang, Xucong, Yusuke Sugano, Mario Fritz, and Andreas Bulling. "Appearance-based Gaze Estimation in the Wild." Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015. arXiv:1504.02863, Project Page
Zhang, Xucong, Yusuke Sugano, Mario Fritz, and Andreas Bulling. "It's Written All Over Your Face: Full-Face Appearance-Based Gaze Estimation." Proc. of the IEEE Conference on Computer Vision and Pattern Recognition Workshops(CVPRW), 2017. arXiv:1611.08860, Project Page
Zhang, Xucong, Yusuke Sugano, Mario Fritz, and Andreas Bulling. "MPIIGaze: Real-World Dataset and Deep Appearance-Based Gaze Estimation." IEEE transactions on pattern analysis and machine intelligence 41 (2017). arXiv:1711.09017
Zhang, Xucong, Yusuke Sugano, and Andreas Bulling. "Evaluation of Appearance-Based Methods and Implications for Gaze-Based Applications." Proc. ACM SIGCHI Conference on Human Factors in Computing Systems (CHI), 2019. arXiv, code