TFLite support for architectures other than SSD in TensorFlow 2 Detection Model Zoo

1. The entire URL of the file you are using

https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/running_on_mobile_tf2.md

2. Describe the feature you request

TFLite support for architectures other than SSD in TensorFlow 2 Detection Model Zoo. There are 4 types of models in the model zoo. The documents says that currently TFLite only supports SSD. Hopefully you can provide support for Faster RCNN, CenterNet and EfficientDet.

3. Additional context

I wonder what's the TFLite support for any other models that's not in the model zoo? Is there a general rule of thumb for what kind of models are supported, and what models are not?