Yinglin Zheng

Hi, thanks for your attention. The pretrained backbones we released contain the weights of the text encoder. In fact, you can load the weights of FaRL using exactly the same...

Hi, Thank you for your attention, 1. We do plan to release the LAION-Face subset(metadata) for further research, please stay tuned. 2. No, we haven't, as face recognition is not...

Hi, we have release the LAION-Face dataset, check it out. https://github.com/FacePerceiver/LAION-Face

Hi, you can check out https://github.com/FacePerceiver/facer for downstream usage, we will continuously update the pre-trained downstream face models in here.

Face alignment model trained on IBUG300W, AFLW19, WFLW dataset is available, check it out [here](https://github.com/FacePerceiver/facer/blob/main/samples/face_alignment.ipynb).

Face alignment model trained on IBUG300W, AFLW19, WFLW dataset is available, check it out [here](https://github.com/FacePerceiver/facer/blob/main/samples/face_alignment.ipynb).

Thank your for the code snippet, this behavior now is better supported in `pyfacer==0.0.5` ```python import torch.nn.functional as F # on already face aligned image, for example FFHQ dataset or...