Open2021/12/18にコメント追加3

MBLLEN のエクスポート

TensorFlow

Keras

ONNX

PINTO

docker run --rm -it --gpus all \
-v `pwd`:/home/user/workdir \
pinto0309/whenet:latest /bin/bash

cd /home/user/workdir/main

test.py

from glob import glob
import numpy as np
import scipy
import keras
import os
import Network
import utls
import time
import cv2
import argparse

"""
python3 test.py -m LOL_img_lowlight
python3 test.py -m Syn_img_lowlight
python3 test.py -m Syn_img_lowlight_withnoise
"""


parser = argparse.ArgumentParser()
#parser.add_argument("--input", "-i", type=str, default='../input', help='test image folder')
#parser.add_argument("--result", "-r", type=str, default='../result', help='result folder')
parser.add_argument("--model", "-m", type=str, default='Syn_img_lowlight_withnoise', help='model name')
#parser.add_argument("--com", "-c", type=int, default=1, help='output with/without origional image and mid result')
#parser.add_argument("--highpercent", "-hp", type=int, default=95, help='should be in [85,100], linear amplification')
#parser.add_argument("--lowpercent", "-lp", type=int, default=5, help='should be in [0,15], rescale the range [p%,1] to [0, 1]')
#parser.add_argument("--gamma", "-g", type=int, default=8, help='should be in [6,10], increase the saturability')
#parser.add_argument("--maxrange", "-mr", type=int, default=8, help='linear amplification range')
arg = parser.parse_args()

# result_folder = arg.result
# if not os.path.isdir(result_folder):
#     os.makedirs(result_folder)

# input_folder = arg.input
# path = glob(input_folder+'/*.*')
# opt = keras.optimizers.Adam(lr=2 * 1e-04, beta_1=0.9, beta_2=0.999, epsilon=1e-08)
# mbllen.compile(loss='mse', optimizer=opt)

import tensorflow as tf
from keras.models import load_model
from keras import backend as K
from tensorflow.python.framework import graph_io
from tensorflow.python.framework import graph_util
def freeze_graph(session, output, save_pb_dir='.', save_pb_name='frozen_graph.pb', save_pb_as_text=False):
    graph = session.graph
    with graph.as_default():
        graphdef_inf = tf.graph_util.remove_training_nodes(graph.as_graph_def())
        print(graphdef_inf)
        graphdef_frozen = graph_util.convert_variables_to_constants(session, graphdef_inf, output)
        graph_io.write_graph(graphdef_frozen, save_pb_dir, save_pb_name, as_text=save_pb_as_text)
        return graphdef_frozen
K.clear_session()
K.set_learning_phase(0)
session = K.get_session()
model_name = arg.model
mbllen = Network.build_mbllen((None, None, 3))
mbllen.load_weights('../models/'+model_name+'.h5')
mbllen.summary()

for l in mbllen.layers:
    print(l.name)

freeze_graph(session, [out.op.name for out in mbllen.outputs], save_pb_dir='.', save_pb_name=model_name+'.pb')

import sys
sys.exit(0)








flag = arg.com
lowpercent = arg.lowpercent
highpercent = arg.highpercent
maxrange = arg.maxrange/10.
hsvgamma = arg.gamma/10.
for i in range(len(path)):
    img_A_path = path[i]
    img_A = utls.imread_color(img_A_path)
    img_A = img_A[np.newaxis, :]

    starttime = time.clock()
    out_pred = mbllen.predict(img_A)
    endtime = time.clock()
    print('The ' + str(i+1)+'th image\'s Time:' +str(endtime-starttime)+'s.')
    fake_B = out_pred[0, :, :, :3]
    fake_B_o = fake_B

    gray_fake_B = fake_B[:, :, 0] * 0.299 + fake_B[:, :, 1] * 0.587 + fake_B[:, :, 1] * 0.114
    percent_max = sum(sum(gray_fake_B >= maxrange))/sum(sum(gray_fake_B <= 1.0))
    # print(percent_max)
    max_value = np.percentile(gray_fake_B[:], highpercent)
    if percent_max < (100-highpercent)/100.:
        scale = maxrange / max_value
        fake_B = fake_B * scale
        fake_B = np.minimum(fake_B, 1.0)

    gray_fake_B = fake_B[:,:,0]*0.299 + fake_B[:,:,1]*0.587 + fake_B[:,:,1]*0.114
    sub_value = np.percentile(gray_fake_B[:], lowpercent)
    fake_B = (fake_B - sub_value)*(1./(1-sub_value))

    imgHSV = cv2.cvtColor(fake_B, cv2.COLOR_RGB2HSV)
    H, S, V = cv2.split(imgHSV)
    S = np.power(S, hsvgamma)
    imgHSV = cv2.merge([H, S, V])
    fake_B = cv2.cvtColor(imgHSV, cv2.COLOR_HSV2RGB)
    fake_B = np.minimum(fake_B, 1.0)

    if flag:
        outputs = np.concatenate([img_A[0,:,:,:], fake_B_o, fake_B], axis=1)
    else:
        outputs = fake_B

    filename = os.path.basename(path[i])
    img_name = result_folder+'/' + filename
    # scipy.misc.toimage(outputs * 255, high=255, low=0, cmin=0, cmax=255).save(img_name)
    outputs = np.minimum(outputs, 1.0)
    outputs = np.maximum(outputs, 0.0)
    utls.imwrite(img_name, outputs)

python3 test.py -m LOL_img_lowlight
python3 test.py -m Syn_img_lowlight
python3 test.py -m Syn_img_lowlight_withnoise

PINTO

exit

sudo chown $USER main/*.pb

PINTO

docker run --gpus all -it --rm \
-v `pwd`:/home/user/workdir \
ghcr.io/pinto0309/openvino2tensorflow:latest

rm -rf openvino
cd main

MODEL=LOL_img_lowlight
H=180
W=320
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32
H=240
W=320
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32
H=360
W=640
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32
H=480
W=640
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32
H=720
W=1280
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32

MODEL=Syn_img_lowlight
H=180
W=320
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32
H=240
W=320
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32
H=360
W=640
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32
H=480
W=640
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32
H=720
W=1280
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32

MODEL=Syn_img_lowlight_withnoise
H=180
W=320
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32
H=240
W=320
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32
H=360
W=640
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32
H=480
W=640
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32
H=720
W=1280
python3 $INTEL_OPENVINO_DIR/deployment_tools/model_optimizer/mo_tf.py \
--input_model ${MODEL}.pb \
--input_shape [1,${H},${W},3] \
--data_type FP32 \
--output_dir openvino/${MODEL}_${H}x${W}/FP32