EmotionUI
Software for Multimodalty 2D+3D Facial Expression Recognition (FER) UI.
demo screenshot
(with RealSense)
required packages
- Python >= 3.6
- numpy >= 1.19.5
- Opencv-python >= 4.5
- PySide6 >= 6.2.1
- PyTorch >= 1.10
- TorchVision >= 0.11
hardware
- usb camera
- Intel RealSense (optional for depth imaging)
- NV GPU (optional, CPU version also works)
loding weights
Two ways, try them, depending on your Internet speed.
-
Manually download. Download vgg16 weights from:
https://drive.google.com/file/d/1f-tKgovJ54l9xR3oIZ6gy77NdPirUddr/view?usp=sharing
Then, move the weights to "./weights" folder, and rename it with "ui_weights.pth".
or
- Run script. Open terminal, run:
> cd weights > python download_weights.py
train your own model
> cd train/dataset
> python main_train.py
Remember to move the new weights to "weights" folder, and rename it with "ui_weights.pth".
notice
The current FER network is vgg16 for simplicity. Current prediction is based on 2D result. One can easily change the network as you wish.