I'm using the BlazeFace model from TensorFlow which only has this few keypoints:
I need those keypoints plus a head keypoint, like this one:
My question is, which would be the best way to approximate the head point (without retraining the model, because I don't have the data that was used for this) having only those face keypoints?