num_epochs=1)
image, annotation = read_tfrecord_and_decode_into_image_annotation_pair_tensors(
filename_queue)
# Fake batch for image and annotation by adding
# leading empty axis.
image_batch_tensor = tf.expand_dims(image, axis=0)
annotation_batch_tensor = tf.expand_dims(annotation, axis=0)
# Be careful: after adaptation, network returns final labels
# and not logits
FCN_8s = adapt_network_for_any_size_input(FCN_8s, 32)
pred, fcn_16s_variables_mapping = FCN_8s(image_batch_tensor=image_batch_tensor,
number_of_classes=number_of_classes,
is_training=False)
# Take away the masked out values from evaluation
weights = tf.to_float(tf.not_equal(annotation_batch_tensor, 255))
# Define the accuracy metric: Mean Intersection Over Union
miou, update_op = slim.metrics.streaming_mean_iou(
predictions=pred,
labels=annotation_batch_tensor,
num_classes=number_of_classes,
weights=weights)
# The op for initializing the variables.
initializer = tf.local_variables_initializer()
# image = distort_randomly_image_color(image)
resized_image, resized_annotation = scale_randomly_image_with_annotation_with_fixed_size_output(
image, annotation, image_train_size)
resized_annotation = tf.squeeze(resized_annotation)
image_batch, annotation_batch = tf.train.shuffle_batch(
[resized_image, resized_annotation],
batch_size=1,
capacity=3000,
num_threads=2,
min_after_dequeue=1000)
upsampled_logits_batch, fcn_16s_variables_mapping = FCN_8s(
image_batch_tensor=image_batch,
number_of_classes=number_of_classes,
is_training=True)
valid_labels_batch_tensor, valid_logits_batch_tensor = get_valid_logits_and_labels(
annotation_batch_tensor=annotation_batch,
logits_batch_tensor=upsampled_logits_batch,
class_labels=class_labels)
cross_entropies = tf.nn.softmax_cross_entropy_with_logits(
logits=valid_logits_batch_tensor, labels=valid_labels_batch_tensor)
#cross_entropy_sum = tf.reduce_sum(cross_entropies)
cross_entropy_sum = tf.reduce_mean(cross_entropies)
pred = tf.argmax(upsampled_logits_batch, dimension=3)
def main(argv):
sys.path.append("./tf-image-segmentation/")
sys.path.append("./models/slim/")
os.environ["CUDA_VISIBLE_DEVICES"] = '1'
import logging
logging.basicConfig(filename='log_examp.log', level=logging.INFO)
logging.debug('This message should go to the log file')
logging.info('So should this')
logging.warning('And this, too')
from tf_image_segmentation.models.fcn_8s import FCN_8s
from tf_image_segmentation.utils.inference import adapt_network_for_any_size_input
number_of_classes = 21
image_filename = argv[1]
image_filename_placeholder = tf.placeholder(tf.string)
feed_dict_to_use = {image_filename_placeholder: image_filename}
image_tensor = tf.read_file(image_filename_placeholder)
image_tensor = tf.image.decode_jpeg(image_tensor, channels=3)
image_batch_tensor = tf.expand_dims(image_tensor, axis=0)
FCN_8s = adapt_network_for_any_size_input(FCN_8s, 32)
pred, fcn_16s_variables_mapping = FCN_8s(
image_batch_tensor=image_batch_tensor,
number_of_classes=number_of_classes,
is_training=False)
initializer = tf.local_variables_initializer()
saver = tf.train.Saver()
with tf.Session() as sess:
sess.run(initializer)
saver.restore(sess, './fcn_8s_checkpoint/model_fcn8s_final.ckpt')
image_np, pred_np = sess.run([image_tensor, pred],
feed_dict=feed_dict_to_use)
# io.imshow(image_np)
# io.show()
#
# io.imshow(pred_np.squeeze())
# io.show()
import skimage.morphology
imageSegmentationSign = int(argv[2])
prediction_mask = (pred_np.squeeze() == imageSegmentationSign)
cropped_object = image_np * np.dstack((prediction_mask, ) * 3)
square = skimage.morphology.square(5)
temp = skimage.morphology.binary_erosion(prediction_mask, square)
negative_mask = (temp != True)
eroding_countour = negative_mask * prediction_mask
eroding_countour_img = np.dstack((eroding_countour, ) * 3)
cropped_object[eroding_countour_img] = 248
png_transparancy_mask = np.uint8(prediction_mask * 255)
image_shape = cropped_object.shape
png_array = np.zeros(shape=[image_shape[0], image_shape[1], 4],
dtype=np.uint8)
png_array[:, :, :3] = cropped_object
png_array[:, :, 3] = png_transparancy_mask
# io.imshow(cropped_object)
nowTime = time.strftime('%Y_%m_%d_%H%M%S', time.localtime(time.time()))
io.imsave('image_' + argv[3], png_array)
slim = tf.contrib.slim
from tf_image_segmentation.models.fcn_8s import FCN_8s
from tf_image_segmentation.utils.pascal_voc import pascal_segmentation_lut
from tf_image_segmentation.utils.tf_records import read_tfrecord_and_decode_into_image_annotation_pair_tensors
from tf_image_segmentation.utils.inference import adapt_network_for_any_size_input
# from tf_image_segmentation.utils.visualization import visualize_segmentation_adaptive
pascal_voc_lut = pascal_segmentation_lut()
number_of_classes = 21
image_holder = tf.placeholder(tf.float32, [None, None, None, None],
name='input')
logits, _ = FCN_8s(image_batch_tensor=image_holder,
number_of_classes=number_of_classes,
is_training=False)
saver = tf.train.Saver()
with tf.Session() as sess:
saver.restore(sess, FLAGS.ckpt_dir + "/model_fcn8s_final.ckpt")
print("Export the Model...")
graph_def = sess.graph.as_graph_def()
# import ipdb
# ipdb.set_trace()
freeze_graph_def = graph_util.convert_variables_to_constants(
sess, graph_def, ["fcn_8s/prediction"])
with open(FLAGS.output_file, 'wb') as f:
def setup_train8(self, lr):
"""
Setups queues and model evaluation.
"""
image_batch, annotation_batch = tf.train.shuffle_batch(
[self.resized_image, self.resized_annotation],
batch_size=1,
capacity=3000,
num_threads=2,
min_after_dequeue=1000)
upsampled_logits_batch, fcn_8s_variables_mapping = FCN_8s(
image_batch_tensor=image_batch,
number_of_classes=self.num_labels,
is_training=True)
valid_labels_batch_tensor, valid_logits_batch_tensor = get_valid_logits_and_labels(
annotation_batch_tensor=annotation_batch,
logits_batch_tensor=upsampled_logits_batch,
class_labels=self.class_labels)
# Count true positives, true negatives, false positives and false negatives.
actual = tf.contrib.layers.flatten(tf.cast(annotation_batch, tf.int64))
self.predicted_img = tf.argmax(upsampled_logits_batch, axis=3)
cross_entropies = tf.nn.softmax_cross_entropy_with_logits(
logits=valid_logits_batch_tensor, labels=valid_labels_batch_tensor)
self.cross_entropy_sum = tf.reduce_mean(cross_entropies)
pred = tf.argmax(upsampled_logits_batch, dimension=3)
probabilities = tf.nn.softmax(upsampled_logits_batch)
with tf.variable_scope("adam_vars"):
self.train_step = tf.train.AdamOptimizer(
learning_rate=lr).minimize(self.cross_entropy_sum)
# Variable's initialization functions
self.init_fn = slim.assign_from_checkpoint_fn(
model_path=self.fcn_8s_checkpoint_path,
var_list=fcn_8s_variables_mapping)
global_vars_init_op = tf.global_variables_initializer()
self.merged_summary_op = tf.summary.merge_all()
self.summary_string_writer = tf.summary.FileWriter(
"./log_8_{}".format(lr))
if not os.path.exists(self.checkpointnew):
os.makedirs(self.checkpointnew)
#The op for initializing the variables.
local_vars_init_op = tf.local_variables_initializer()
self.combined_op = tf.group(local_vars_init_op, global_vars_init_op)
# We need this to save only model variables and omit
# optimization-related and other variables.
model_variables = slim.get_model_variables()
self.saver = tf.train.Saver(model_variables)