Converting DeepLabV3 to CoreML Model.

DeepLab is a state-of-art deep learning model for semantic image segmentation.

You can use this model in your iOS mobile devices through converting to CoreML Model.

1. Download the original TensorFlow model as a “frozen_inference_graph.pb” from GitHub.

tensorflow/models

2.Install tfcoreml that convert TensorFlow model to CoreML model.

pip install tfcoreml
import tfcoreml

3.Run convert script.

tfcoreml.convert(tf_model_path = 'frozen_inference_graph.pb',
                 mlmodel_path = 'deeplab.mlmodel',
                 input_name_shape_dict =  {'ImageTensor:0' : [1, 513, 513, 3]},
                 image_input_names=['ImageTensor__0'],
                 output_feature_names = ['ResizeBilinear_2:0'],
                 )

Voila!

Please follow my Twitter. https://twitter.com/JackdeS11 And please clap your hands 👏.