def create_dataset_batch_queue(dataset):
from preprocessing import ssd_vgg_preprocessing
with tf.device('/cpu:0'):
with tf.name_scope(FLAGS.dataset_name + '_data_provider'):
provider = slim.dataset_data_provider.DatasetDataProvider(
dataset,
num_readers=FLAGS.num_readers,
common_queue_capacity=1000 * config.batch_size,
common_queue_min=700 * config.batch_size,
shuffle=True)
# Get for SSD network: image, labels, bboxes.
[image, glabel, gbboxes, x1, x2, x3, x4, y1, y2, y3, y4] = provider.get([
'image',
'object/label',
'object/bbox',
'object/oriented_bbox/x1',
'object/oriented_bbox/x2',
'object/oriented_bbox/x3',
'object/oriented_bbox/x4',
'object/oriented_bbox/y1',
'object/oriented_bbox/y2',
'object/oriented_bbox/y3',
'object/oriented_bbox/y4'
])
gxs = tf.transpose(tf.stack([x1, x2, x3, x4])) #shape = (N, 4)
gys = tf.transpose(tf.stack([y1, y2, y3, y4]))
image = tf.identity(image, 'input_image')
# Pre-processing image, labels and bboxes.
image, glabel, gbboxes, gxs, gys = \
ssd_vgg_preprocessing.preprocess_image(
image, glabel, gbboxes, gxs, gys,
out_shape = config.train_image_shape,
data_format = config.data_format,
use_rotation = config.use_rotation,
is_training = True)
image = tf.identity(image, 'processed_image')
# calculate ground truth
pixel_cls_label, pixel_cls_weight, \
pixel_link_label, pixel_link_weight = \
pixel_link.tf_cal_gt_for_single_image(gxs, gys, glabel)
# batch them
with tf.name_scope(FLAGS.dataset_name + '_batch'):
b_image, b_pixel_cls_label, b_pixel_cls_weight, \
b_pixel_link_label, b_pixel_link_weight = \
tf.train.batch(
[image, pixel_cls_label, pixel_cls_weight,
pixel_link_label, pixel_link_weight],
batch_size = config.batch_size_per_gpu,
num_threads= FLAGS.num_preprocessing_threads,
capacity = 500)
with tf.name_scope(FLAGS.dataset_name + '_prefetch_queue'):
batch_queue = slim.prefetch_queue.prefetch_queue(
[b_image, b_pixel_cls_label, b_pixel_cls_weight,
b_pixel_link_label, b_pixel_link_weight],
capacity = 50)
return batch_queue
def create_dataset_batch_queue(dataset):
from preprocessing import ssd_vgg_preprocessing
with tf.device('/cpu:0'):
with tf.name_scope(FLAGS.dataset_name + '_data_provider'):
provider = slim.dataset_data_provider.DatasetDataProvider(
dataset,
num_readers=FLAGS.num_readers,
common_queue_capacity=1000 * config.batch_size,
common_queue_min=700 * config.batch_size,
shuffle=True)
# Get for SSD network: image, labels, bboxes.
[image, glabel, gbboxes, x1, x2, x3, x4, y1, y2, y3, y4] = provider.get([
'image',
'object/label',
'object/bbox',
'object/oriented_bbox/x1',
'object/oriented_bbox/x2',
'object/oriented_bbox/x3',
'object/oriented_bbox/x4',
'object/oriented_bbox/y1',
'object/oriented_bbox/y2',
'object/oriented_bbox/y3',
'object/oriented_bbox/y4'
])
gxs = tf.transpose(tf.stack([x1, x2, x3, x4])) #shape = (N, 4)
gys = tf.transpose(tf.stack([y1, y2, y3, y4]))
image = tf.identity(image, 'input_image')
# Pre-processing image, labels and bboxes.
image, glabel, gbboxes, gxs, gys = \
ssd_vgg_preprocessing.preprocess_image(
image, glabel, gbboxes, gxs, gys,
out_shape = config.train_image_shape,
data_format = config.data_format,
use_rotation = config.use_rotation,
is_training = True)
image = tf.identity(image, 'processed_image')
# calculate ground truth
pixel_cls_label, pixel_cls_weight, \
pixel_link_label, pixel_link_weight = \
pixel_link.tf_cal_gt_for_single_image(gxs, gys, glabel)
# batch them
with tf.name_scope(FLAGS.dataset_name + '_batch'):
b_image, b_pixel_cls_label, b_pixel_cls_weight, \
b_pixel_link_label, b_pixel_link_weight = \
tf.train.batch(
[image, pixel_cls_label, pixel_cls_weight,
pixel_link_label, pixel_link_weight],
batch_size = config.batch_size_per_gpu,
num_threads= FLAGS.num_preprocessing_threads,
capacity = 500)
with tf.name_scope(FLAGS.dataset_name + '_prefetch_queue'):
batch_queue = slim.prefetch_queue.prefetch_queue(
[b_image, b_pixel_cls_label, b_pixel_cls_weight,
b_pixel_link_label, b_pixel_link_weight],
capacity = 50)
return batch_queue
def create_dataset_batch_queue_multiphase_multislice_mask(dataset):
from preprocessing import ssd_vgg_preprocessing_multiphase_multislice_mask
with tf.device('/cpu:0'):
with tf.name_scope(FLAGS.dataset_name + '_data_provider'):
provider = slim.dataset_data_provider.DatasetDataProvider(
dataset,
num_readers=FLAGS.num_readers,
common_queue_capacity=1000 * config.batch_size,
common_queue_min=700 * config.batch_size,
shuffle=True)
[nc_image, art_image, pv_image, mask_image, shape, glabel, gbboxes, x1, x2, x3, x4, y1, y2, y3, y4] = provider.get(
[
'nc_image', 'art_image',
'pv_image', 'mask_image', 'shape',
'object/label',
'object/bbox',
'object/oriented_bbox/x1',
'object/oriented_bbox/x2',
'object/oriented_bbox/x3',
'object/oriented_bbox/x4',
'object/oriented_bbox/y1',
'object/oriented_bbox/y2',
'object/oriented_bbox/y3',
'object/oriented_bbox/y4'])
gxs = tf.transpose(tf.stack([x1, x2, x3, x4])) # shape = (N, 4)
gys = tf.transpose(tf.stack([y1, y2, y3, y4]))
nc_image = tf.identity(nc_image, 'input_nc_image')
art_image = tf.identity(art_image, 'input_art_image')
pv_image = tf.identity(pv_image, 'input_pv_image')
mask_image = tf.identity(mask_image, 'mask_image')
# Pre-processing image, labels and bboxes.
print('nc image: ', nc_image)
print('art image: ', art_image)
print('pv image: ', pv_image)
print('mask image: ', mask_image)
# mask should be processed with original image at same time
nc_image, art_image, pv_image, mask_image, glabel, gbboxes, gxs, gys = \
ssd_vgg_preprocessing_multiphase_multislice_mask.preprocess_image_multiphase_multislice_mask(
nc_image, art_image, pv_image, mask_image, glabel, gbboxes, gxs, gys,
out_shape=config.train_image_shape,
data_format=config.data_format,
use_rotation=config.use_rotation,
is_training=True)
nc_image = tf.identity(nc_image, 'processed_nc_image')
art_image = tf.identity(art_image, 'processed_art_image')
pv_image = tf.identity(pv_image, 'processed_pv_image')
mask_image = tf.identity(mask_image, 'processed_mask_image')
# calculate ground truth
pixel_cls_label, pixel_cls_weight, \
pixel_link_label, pixel_link_weight = \
pixel_link.tf_cal_gt_for_single_image(gxs, gys, glabel)
# batch them
with tf.name_scope(FLAGS.dataset_name + '_batch'):
b_nc_image, b_art_image, b_pv_image, b_mask_image, b_pixel_cls_label, b_pixel_cls_weight, \
b_pixel_link_label, b_pixel_link_weight = \
tf.train.batch(
[nc_image, art_image, pv_image, mask_image, pixel_cls_label, pixel_cls_weight,
pixel_link_label, pixel_link_weight],
batch_size=config.batch_size_per_gpu,
num_threads=FLAGS.num_preprocessing_threads,
capacity=500)
with tf.name_scope(FLAGS.dataset_name + '_prefetch_queue'):
batch_queue = slim.prefetch_queue.prefetch_queue(
[b_nc_image, b_art_image, b_pv_image, b_mask_image, b_pixel_cls_label, b_pixel_cls_weight,
b_pixel_link_label, b_pixel_link_weight],
capacity=50)
return batch_queue