-
Notifications
You must be signed in to change notification settings - Fork 0
/
train_inputs_precomputed.py
172 lines (137 loc) · 6.9 KB
/
train_inputs_precomputed.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
"""
The heatmaps have been precomputed and saved in the tfrecord file.
"""
import numpy as np
import tensorflow as tf
from inputs import distort_color, apply_with_random_selector, flip_parts_left_right, flip_heatmaps_left_right
def input_nodes(
tfrecords,
num_parts,
# number of times to read the tfrecords
num_epochs=None,
# Data queue feeding the model
batch_size=32,
num_threads=2,
shuffle_batch = True,
capacity = 1000,
min_after_dequeue = 96,
# Tensorboard Summaries
add_summaries = True,
# Global configuration
cfg=None):
with tf.name_scope('inputs'):
# A producer to generate tfrecord file paths
filename_queue = tf.train.string_input_producer(
tfrecords,
num_epochs=num_epochs,
shuffle=shuffle_batch
)
# Construct a Reader to read examples from the tfrecords file
reader = tf.TFRecordReader()
_, serialized_example = reader.read(filename_queue)
# Parse an Example to access the Features
features = tf.parse_single_example(
serialized_example,
features = {
'image/id' : tf.FixedLenFeature([], tf.string),
'image/encoded' : tf.FixedLenFeature([], tf.string),
'image/height' : tf.FixedLenFeature([], tf.int64),
'image/width' : tf.FixedLenFeature([], tf.int64),
'image/object/bbox/xmin' : tf.VarLenFeature(dtype=tf.float32),
'image/object/bbox/ymin' : tf.VarLenFeature(dtype=tf.float32),
'image/object/bbox/xmax' : tf.VarLenFeature(dtype=tf.float32),
'image/object/bbox/ymax' : tf.VarLenFeature(dtype=tf.float32),
'image/object/parts/x' : tf.VarLenFeature(dtype=tf.float32), # x coord for all parts and all objects
'image/object/parts/y' : tf.VarLenFeature(dtype=tf.float32), # y coord for all parts and all objects
'image/object/parts/v' : tf.VarLenFeature(dtype=tf.int64), # part visibility for all parts and all objects
'image/object/parts/heatmaps' : tf.FixedLenFeature([cfg.HEATMAP_SIZE * cfg.HEATMAP_SIZE * num_parts], dtype=tf.float32)
}
)
# Read in a jpeg image
image = tf.image.decode_jpeg(features['image/encoded'], channels=3)
# Convert the pixel values to be in the range [0,1]
if image.dtype != tf.float32:
image = tf.image.convert_image_dtype(image, dtype=tf.float32)
image_height = tf.cast(features['image/height'], tf.float32)
image_width = tf.cast(features['image/width'], tf.float32)
image_id = features['image/id']
xmin = tf.expand_dims(features['image/object/bbox/xmin'].values, 0)
ymin = tf.expand_dims(features['image/object/bbox/ymin'].values, 0)
xmax = tf.expand_dims(features['image/object/bbox/xmax'].values, 0)
ymax = tf.expand_dims(features['image/object/bbox/ymax'].values, 0)
parts_x = tf.expand_dims(features['image/object/parts/x'].values, 0)
parts_y = tf.expand_dims(features['image/object/parts/y'].values, 0)
parts_v = tf.cast(tf.expand_dims(features['image/object/parts/v'].values, 0), tf.int32)
heatmaps = features['image/object/parts/heatmaps']
heatmaps = tf.reshape(heatmaps, [cfg.HEATMAP_SIZE, cfg.HEATMAP_SIZE, num_parts])
# Add a summary of the original data
if add_summaries:
bboxes_to_draw = tf.transpose(tf.concat(0, [ymin, xmin, ymax, xmax]), [1, 0])
bboxes_to_draw = tf.reshape(bboxes_to_draw, [1, -1, 4])
image_with_bboxes = tf.image.draw_bounding_boxes(tf.expand_dims(image, 0), bboxes_to_draw)
tf.image_summary('original_image', image_with_bboxes)
# Randomly flip the image:
if cfg.DO_RANDOM_FLIP_LEFT_RIGHT:
r = tf.random_uniform([], minval=0, maxval=1, dtype=tf.float32)
do_flip = tf.less(r, 0.5)
image = tf.cond(do_flip, lambda: tf.image.flip_left_right(image), lambda: tf.identity(image))
xmin, xmax = tf.cond(do_flip, lambda: tf.tuple([1. - xmax, 1. - xmin]), lambda: tf.tuple([xmin, xmax]))
parts_x, parts_y, parts_v = tf.cond(do_flip,
lambda: tf.py_func(flip_parts_left_right, [parts_x, parts_y, parts_v, cfg.PARTS.LEFT_RIGHT_PAIRS, num_parts], [tf.float32, tf.float32, tf.int32]),
lambda: tf.tuple([parts_x, parts_y, parts_v])
)
heatmaps = tf.cond(do_flip,
lambda: tf.py_func(flip_heatmaps_left_right, [heatmaps, cfg.PARTS.LEFT_RIGHT_PAIRS], [tf.float32]),
lambda: tf.identity(heatmaps)
)
parts = tf.reshape(tf.transpose(tf.concat(0, [parts_x, parts_y])), [-1])
part_visibilities = tf.reshape(parts_v, [-1])# tf.reshape(parts_v, tf.pack([num_bboxes, num_parts]))
# Distort the colors
r = tf.random_uniform([], minval=0, maxval=1, dtype=tf.float32)
do_color_distortion = tf.less(r, cfg.DO_COLOR_DISTORTION)
num_color_cases = 1 if cfg.COLOR_DISTORT_FAST else 4
distorted_image = apply_with_random_selector(
image,
lambda x, ordering: distort_color(x, ordering, fast_mode=cfg.COLOR_DISTORT_FAST),
num_cases=num_color_cases)
image = tf.cond(do_color_distortion, lambda: tf.identity(distorted_image), lambda: tf.identity(image))
image.set_shape([cfg.INPUT_SIZE, cfg.INPUT_SIZE, 3])
# Add a summary
if add_summaries:
bboxes_to_draw = tf.transpose(tf.concat(0, [ymin, xmin, ymax, xmax]), [1, 0])
bboxes_to_draw = tf.reshape(bboxes_to_draw, [1, -1, 4])
image_with_bboxes = tf.image.draw_bounding_boxes(tf.expand_dims(image, 0), bboxes_to_draw)
tf.image_summary('flipped_distorted_image', image_with_bboxes)
image = tf.sub(image, 0.5)
image = tf.mul(image, 2.0)
# Set the shape of everything for the queue
image.set_shape([cfg.INPUT_SIZE, cfg.INPUT_SIZE, 3])
image_ids = [image_id]
heatmaps.set_shape([cfg.HEATMAP_SIZE, cfg.HEATMAP_SIZE, num_parts])
bboxes = tf.concat(0, [xmin, ymin, xmax, ymax])
bboxes = tf.reshape(tf.transpose(bboxes, [1, 0]), [4])
bboxes.set_shape([4])
parts.set_shape([num_parts * 2])
part_visibilities.set_shape([num_parts])
if shuffle_batch:
batched_images, batched_heatmaps, batched_parts, batched_part_visibilities, batched_image_ids = tf.train.shuffle_batch(
[image, heatmaps, parts, part_visibilities, image_ids],
batch_size=batch_size,
num_threads=num_threads,
capacity= capacity, #batch_size * (num_threads + 2),
# Ensures a minimum amount of shuffling of examples.
min_after_dequeue= min_after_dequeue, # 3 * batch_size,
seed = cfg.RANDOM_SEED,
enqueue_many=False,
name="shuffle_batch_queue"
)
else:
batched_images, batched_heatmaps, batched_parts, batched_part_visibilities, batched_image_ids = tf.train.batch(
[image, heatmaps, parts, part_visibilities, image_ids],
batch_size=batch_size,
num_threads=num_threads,
capacity= capacity, #batch_size * (num_threads + 2),
enqueue_many=False
)
# return a batch of images and their labels
return batched_images, batched_heatmaps, batched_parts, batched_part_visibilities, batched_image_ids