Convert BlazeFace .tflite to .pb

blazeface_tflite_to_pb.md

This script is used to convert BlazeFace tflite model to pb.

you need to get below files first.

• model: https://github.com/google/mediapipe/blob/master/mediapipe/models/face_detection_front.tflite
• flatc: https://github.com/tensorflow/tensorflow/blob/v1.14.0/tensorflow/lite/tools/visualize.py#L38
• shema: https://github.com/tensorflow/tensorflow/blob/v1.14.0/tensorflow/lite/tools/visualize.py#L37

blazeface_tflite_to_pb.py

#!/usr/bin/env python
# coding: utf-8

import os
import numpy as np
import json
import tensorflow as tf

os.environ['CUDA_VISIBLE_DEVICES'] = '0'
schema = "schema.fbs"
binary = "flatc"
model_path = "mediapipe/models/facedetector_front.tflite"
output_pb_path = "facedetector_front.pb"
output_savedmodel_path = "saved_model"
model_json_path = "/tmp/facedetector_front.json"
num_tensors = 176
output_node_names = ['classificators', 'regressors']


def gen_model_json():
    if not os.path.exists(model_json_path):
        cmd = (binary +
               " -t --strict-json --defaults-json -o /tmp {schema} -- {input}".
               format(input=model_path, schema=schema))
        os.system(cmd)


def parse_json():
    j = json.load(open(model_json_path))
    op_types = [v['builtin_code'] for v in j['operator_codes']]
    # print('op types:', op_types)
    ops = j['subgraphs'][0]['operators']
    # print('num of ops:', len(ops))
    return ops, op_types


def make_graph(ops, op_types, interpreter):
    tensors = {}
    input_details = interpreter.get_input_details()
    output_details = interpreter.get_output_details()
    # print(input_details)
    for input_detail in input_details:
        tensors[input_detail['index']] = tf.placeholder(
            dtype=input_detail['dtype'],
            shape=input_detail['shape'],
            name=input_detail['name'])

    for index, op in enumerate(ops):
        print('op: ', op)
        op_type = op_types[op['opcode_index']]
        if op_type == 'CONV_2D':
            input_tensor = tensors[op['inputs'][0]]
            weights_detail = interpreter._get_tensor_details(op['inputs'][1])
            bias_detail = interpreter._get_tensor_details(op['inputs'][2])
            output_detail = interpreter._get_tensor_details(op['outputs'][0])
            # print('weights_detail: ', weights_detail)
            # print('bias_detail: ', bias_detail)
            # print('output_detail: ', output_detail)
            weights_array = interpreter.get_tensor(weights_detail['index'])
            weights_array = np.transpose(weights_array, (1, 2, 3, 0))
            bias_array = interpreter.get_tensor(bias_detail['index'])
            weights = tf.Variable(weights_array, name=weights_detail['name'])
            bias = tf.Variable(bias_array, name=bias_detail['name'])
            options = op['builtin_options']
            output_tensor = tf.nn.conv2d(
                input_tensor,
                weights,
                strides=[1, options['stride_h'], options['stride_w'], 1],
                padding=options['padding'],
                dilations=[
                    1, options['dilation_h_factor'],
                    options['dilation_w_factor'], 1
                ],
                name=output_detail['name'] + '/conv2d')
            output_tensor = tf.add(
                output_tensor, bias, name=output_detail['name'])
            tensors[output_detail['index']] = output_tensor
        elif op_type == 'DEPTHWISE_CONV_2D':
            input_tensor = tensors[op['inputs'][0]]
            weights_detail = interpreter._get_tensor_details(op['inputs'][1])
            bias_detail = interpreter._get_tensor_details(op['inputs'][2])
            output_detail = interpreter._get_tensor_details(op['outputs'][0])
            # print('weights_detail: ', weights_detail)
            # print('bias_detail: ', bias_detail)
            # print('output_detail: ', output_detail)
            weights_array = interpreter.get_tensor(weights_detail['index'])
            weights_array = np.transpose(weights_array, (1, 2, 3, 0))
            bias_array = interpreter.get_tensor(bias_detail['index'])
            weights = tf.Variable(weights_array, name=weights_detail['name'])
            bias = tf.Variable(bias_array, name=bias_detail['name'])
            options = op['builtin_options']
            output_tensor = tf.nn.depthwise_conv2d(
                input_tensor,
                weights,
                strides=[1, options['stride_h'], options['stride_w'], 1],
                padding=options['padding'],
                # dilations=[
                #     1, options['dilation_h_factor'],
                #     options['dilation_w_factor'], 1
                # ],
                name=output_detail['name'] + '/depthwise_conv2d')
            output_tensor = tf.add(
                output_tensor, bias, name=output_detail['name'])
            tensors[output_detail['index']] = output_tensor
        elif op_type == 'MAX_POOL_2D':
            input_tensor = tensors[op['inputs'][0]]
            output_detail = interpreter._get_tensor_details(op['outputs'][0])
            options = op['builtin_options']
            output_tensor = tf.nn.max_pool(
                input_tensor,
                ksize=[
                    1, options['filter_height'], options['filter_width'], 1
                ],
                strides=[1, options['stride_h'], options['stride_w'], 1],
                padding=options['padding'],
                name=output_detail['name'])
            tensors[output_detail['index']] = output_tensor
        elif op_type == 'PAD':
            input_tensor = tensors[op['inputs'][0]]
            output_detail = interpreter._get_tensor_details(op['outputs'][0])
            paddings_detail = interpreter._get_tensor_details(op['inputs'][1])
            # print('output_detail:', output_detail)
            # print('paddings_detail:', paddings_detail)
            paddings_array = interpreter.get_tensor(paddings_detail['index'])
            paddings = tf.Variable(
                paddings_array, name=paddings_detail['name'])
            output_tensor = tf.pad(
                input_tensor, paddings, name=output_detail['name'])
            tensors[output_detail['index']] = output_tensor
        elif op_type == 'RELU':
            output_detail = interpreter._get_tensor_details(op['outputs'][0])
            input_tensor = tensors[op['inputs'][0]]
            output_tensor = tf.nn.relu(
                input_tensor, name=output_detail['name'])
            tensors[output_detail['index']] = output_tensor
        elif op_type == 'RESHAPE':
            input_tensor = tensors[op['inputs'][0]]
            output_detail = interpreter._get_tensor_details(op['outputs'][0])
            options = op['builtin_options']
            output_tensor = tf.reshape(
                input_tensor, options['new_shape'], name=output_detail['name'])
            tensors[output_detail['index']] = output_tensor
        elif op_type == 'ADD':
            output_detail = interpreter._get_tensor_details(op['outputs'][0])
            input_tensor_0 = tensors[op['inputs'][0]]
            input_tensor_1 = tensors[op['inputs'][1]]
            output_tensor = tf.add(
                input_tensor_0, input_tensor_1, name=output_detail['name'])
            tensors[output_detail['index']] = output_tensor
        elif op_type == 'CONCATENATION':
            output_detail = interpreter._get_tensor_details(op['outputs'][0])
            input_tensor_0 = tensors[op['inputs'][0]]
            input_tensor_1 = tensors[op['inputs'][1]]
            options = op['builtin_options']
            output_tensor = tf.concat([input_tensor_0, input_tensor_1],
                                      options['axis'],
                                      name=output_detail['name'])
            tensors[output_detail['index']] = output_tensor
        else:
            raise ValueError(op_type)


def main():
    gen_model_json()
    ops, op_types = parse_json()

    interpreter = tf.lite.Interpreter(model_path)
    interpreter.allocate_tensors()
    # input_details = interpreter.get_input_details()
    # output_details = interpreter.get_output_details()
    # print(input_details)
    # print(output_details)
    # for i in range(num_tensors):
    #     detail = interpreter._get_tensor_details(i)
    #     print(detail)

    make_graph(ops, op_types, interpreter)

    config = tf.ConfigProto()
    config.gpu_options.allow_growth = True
    graph = tf.get_default_graph()
    # writer = tf.summary.FileWriter(os.path.splitext(output_pb_path)[0])
    # writer.add_graph(graph)
    # writer.flush()
    # writer.close()
    with tf.Session(config=config, graph=graph) as sess:
        sess.run(tf.global_variables_initializer())
        graph_def = tf.graph_util.convert_variables_to_constants(
            sess=sess,
            input_graph_def=graph.as_graph_def(),
            output_node_names=output_node_names)

        with tf.gfile.GFile(output_pb_path, 'wb') as f:
            f.write(graph_def.SerializeToString())

        tf.saved_model.simple_save(
            sess,
            output_savedmodel_path,
            inputs={'input': graph.get_tensor_by_name('input:0')},
            outputs={
                'classificators': graph.get_tensor_by_name('classificators:0'),
                'regressors': graph.get_tensor_by_name('regressors:0')
            })


if __name__ == '__main__':
    main()

I'm running into issues parsing the json. There are missing opcode_index attributes:

('op: ', {u'inputs': [0, 1, 2], u'builtin_options': {u'stride_w': 2, u'stride_h': 2}, u'builtin_options_type': u'Conv2DOptions', u'outputs': [3]})
Traceback (most recent call last):
  File "blazeface_tflite_to_pb.py", line 203, in <module>
    main()
  File "blazeface_tflite_to_pb.py", line 173, in main
    make_graph(ops, op_types, interpreter)
  File "blazeface_tflite_to_pb.py", line 46, in make_graph
    op_type = op_types[op['opcode_index']]
KeyError: 'opcode_index'

How should i handle these?

flatc -t --strict-json --defaults-json -o schema.fbs -- face_detection_front.tflite
flatc: error: current schema has no file_identifier: cannot test if "face_detection_front.tflite" matches the schema, use --raw-binary to read this file anyway.

I was able to use this great code example to successfully convert the BlazeFace, FaceMesh and 3D Hand Pose into a saved_model. If it's useful to you, I'm happy. I used Tensorflow-GPU v1.15.2.

・My repository
　https://github.com/PINTO0309/PINTO_model_zoo

・Example

1. BlazeFace
2. FaceMesh
3. Hand Detection and Tracking
4. BlazePose
5. BlazePose (Full Keypoints)
6. KNIFT
7. Iris Landmark
8. Objectron
9. BodyPix

I was able to use this great code example to successfully convert the BlazeFace, FaceMesh and 3D Hand Pose into a saved_model. If it's useful to you, I'm happy. I used Tensorflow-GPU v1.15.2.

・My repository
　https://github.com/PINTO0309/PINTO_model_zoo

・Example

1. BlazeFace
2. FaceMesh
3. Hand Detection and Tracking

@PINTO0309 the links are broken. Can you fix them?

@SocialKitLtd
I fixed it.

@PINTO0309 Thanks! But I can't seems to find the .pb model. Is it there? Or I'm miss understanding what you wrote

@SocialKitLtd Run download.sh or download_new.sh to download all the materials you need.

@PINTO0309 On which of the folders? Sorry, I don't really have deep understand ML yet:

01_float32

@PINTO0309 Works!! :) Katsuya, I sent you an email from [email protected]. Did you get it by chance?

These locations are now not valid-

flatc: https://github.com/tensorflow/tensorflow/blob/release_1.14.0/tensorflow/lite/tools/visualize.py#L38
shema: https://github.com/tensorflow/tensorflow/blob/release_1.14.0/tensorflow/lite/tools/visualize.py#L37

These locations are now not valid-

flatc: https://github.com/tensorflow/tensorflow/blob/release_1.14.0/tensorflow/lite/tools/visualize.py#L38
shema: https://github.com/tensorflow/tensorflow/blob/release_1.14.0/tensorflow/lite/tools/visualize.py#L37

They are still valid at:
flatc: https://github.com/tensorflow/tensorflow/blob/v1.14.0/tensorflow/lite/tools/visualize.py#L38
schema: https://github.com/tensorflow/tensorflow/blob/v1.14.0/tensorflow/lite/tools/visualize.py#L37

RuntimeError: Sorry, schema file cannot be found at '../schema/schema.fbs' . where do i get this file?