Convert the face_landmark.tflite file in your solution back to face_landmark.pb (protobuf file).

I am trying to convert the "face_landmark.tflite" file in mediapipe solution (https://github.com/google-ai-edge/mediapipe) back to face_landmark.pb (protobuf file). I am working with C# code and unable to find any wrapper frameworks which make use of .tflite files for face landmark detection. Tensorflow.NET does not support an interpreter to load the .tflite models. But it supports loading models in .pb or "savedModel" format.
Can you please let me know if you have the original tensorFlow model in 'SavedModel of .pb format for Fare landmark detection(Mediapipe Facemesh)? I can use this with TensorFlow.net.

Also, is any one of the pre-trained model of TensorFlow (link provided below) used for face_landMark.tflite model generation? Please suggest.
https://github.com/tensorflow/models/tree/master/research/slim#Pretrained