-
Notifications
You must be signed in to change notification settings - Fork 17
/
NYU_dataset.py
272 lines (209 loc) · 10 KB
/
NYU_dataset.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
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
"""
Reference: https://github.com/cleinc/bts (From Big to Small: Multi-Scale Local Planar Guidance for Monocular Depth Estimation)
"""
import pandas as pd
import os
from omegaconf import DictConfig
from pytorch_lightning import LightningDataModule
from torch.utils.data import Dataset, DataLoader
from torchvision import transforms
from PIL import Image
import numpy as np
import torch
import random
import scipy.ndimage as ndimage
# from .transforms import *
def _is_pil_image(img):
return isinstance(img, Image.Image)
def _is_numpy_image(img):
return isinstance(img, np.ndarray) and (img.ndim in {2, 3})
def preprocessing_transforms(mode):
return transforms.Compose([
ToTensor(mode=mode)
])
class DataLoadPreprocess(Dataset):
def __init__(self, csv_file, config, mode, transform=None, eval=False):
self.paths = pd.read_csv(csv_file, header=None,
names=['image', 'depth'])
self.config = config
self.mode = mode
self.transform = transform
self.to_tensor = ToTensor
self.dataset_path = os.path.dirname(os.path.dirname(csv_file))
self.input_height = self.config.DATA.TRAIN_CROP_SIZE[1]
self.input_width = self.config.DATA.TRAIN_CROP_SIZE[0]
self.do_random_rotate=True
self.degree = 2.5
def __getitem__(self, idx):
image_path = self.dataset_path + "/"+ self.paths['image'][idx]
depth_path = self.dataset_path + "/"+ self.paths['depth'][idx]
image = Image.open(image_path)
depth_gt = Image.open(depth_path)
depth_gt = depth_gt.crop((41, 45, 601, 471))
image = image.crop((41, 45, 601, 471))
if self.mode == 'train':
depth_completed_path = depth_path #self.dataset_path + '/' + self.paths['depth'][idx].replace("train","train_completed")
depth_completed = Image.open(depth_completed_path)
depth_completed = depth_completed.crop((41, 45, 601, 471))
if self.mode == 'train' and self.do_random_rotate is True:
random_angle = (random.random() - 0.5) * 2 * self.degree
image = self.rotate_image(image, random_angle)
depth_gt = self.rotate_image(depth_gt, random_angle, flag=Image.NEAREST)
depth_completed = self.rotate_image(depth_completed, random_angle, flag=Image.NEAREST)
image = np.asarray(image, dtype=np.float32) / 255.0
depth_gt = np.asarray(depth_gt, dtype=np.float32)
depth_gt = np.expand_dims(depth_gt, axis=2)
if self.mode == 'train':
depth_gt = depth_gt * 1000.0
depth_gt = depth_gt / 255.0
else:
depth_gt = depth_gt / 1000.0
if self.mode == 'train':
depth_completed = np.asarray(depth_completed, dtype=np.float32)
depth_completed = np.expand_dims(depth_completed, axis=2)
depth_completed = depth_completed * 1000.0
depth_completed = depth_completed / 255.0
image, depth_gt, depth_completed = self.random_crop(image, depth_gt, depth_completed, self.input_height, self.input_width)
image, depth_gt, depth_completed = self.train_preprocess(image, depth_completed, depth_gt)
depth_gt = np.clip(depth_gt, 10.0, 1000.0)
depth_gt = 1000. / depth_gt
depth_completed = np.clip(depth_completed, 10.0, 1000.0)
depth_completed = 1000. / depth_completed
sample = {'image': image, 'depth': depth_gt, 'depth_completed': depth_completed}
else:
sample = {'image': image, 'depth': depth_gt}
if self.transform:
sample = self.transform(sample)
if self.mode != 'train':
#print(os.path.basename(self.paths['image'][idx]))
sample['path'] = os.path.basename(self.paths['image'][idx]) #) #.replace("_colors", "")
# sample['path'] = os.path.basename(self.paths['image'][idx]) # ) #.replace("_colors", "")
# print("min image: " + str(torch.min(sample['image']).item()) + " max image: " + str(torch.max(sample['image']).item()))
# print("min depth: " + str(torch.min(sample['depth']).item()) + " max depth: " + str(torch.max(sample['depth']).item()))
return sample
def rotate_image(self, image, angle, flag=Image.BILINEAR):
result = image.rotate(angle, resample=flag)
return result
def random_crop(self, img, depth, depth_completed, height, width):
assert img.shape[0] >= height
assert img.shape[1] >= width
assert img.shape[0] == depth.shape[0]
assert img.shape[1] == depth.shape[1]
assert img.shape[0] == depth_completed.shape[0]
assert img.shape[1] == depth_completed.shape[1]
x = random.randint(0, img.shape[1] - width)
y = random.randint(0, img.shape[0] - height)
img = img[y:y + height, x:x + width, :]
depth = depth[y:y + height, x:x + width, :]
depth_completed = depth_completed[y:y + height, x:x + width, :]
return img, depth, depth_completed
def train_preprocess(self, image, depth_gt, depth_completed):
# Random flipping
do_flip = random.random()
if do_flip > 0.5:
image = (image[:, ::-1, :]).copy()
depth_gt = (depth_gt[:, ::-1, :]).copy()
depth_completed = (depth_completed[:, ::-1, :]).copy()
# Random gamma, brightness, color augmentation
do_augment = random.random()
if do_augment > 0.5:
image = self.augment_image(image)
return image, depth_gt, depth_completed
def augment_image(self, image):
# gamma augmentation
gamma = random.uniform(0.9, 1.1)
image_aug = image ** gamma
# brightness augmentation
brightness = random.uniform(0.75, 1.25)
image_aug = image_aug * brightness
# color augmentation
colors = np.random.uniform(0.9, 1.1, size=3)
white = np.ones((image.shape[0], image.shape[1]))
color_image = np.stack([white * colors[i] for i in range(3)], axis=2)
image_aug *= color_image
image_aug = np.clip(image_aug, 0, 1)
return image_aug
def __len__(self):
#return 500
return len(self.paths)
class ToTensor(object):
def __init__(self, mode):
self.mode = mode
self.normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
def __call__(self, sample):
image = sample['image']
image = self.to_tensor(image)
image = self.normalize(image)
depth = sample['depth']
depth = self.to_tensor(depth)
if 'depth_completed' in sample.keys():
depth_completed = sample['depth_completed']
depth_completed = self.to_tensor(depth_completed)
return {'image': image, 'depth': depth, 'depth_completed':depth_completed}
else:
return {'image': image, 'depth': depth}
def to_tensor(self, pic):
if not (_is_pil_image(pic) or _is_numpy_image(pic)):
raise TypeError(
'pic should be PIL Image or ndarray. Got {}'.format(type(pic)))
if isinstance(pic, np.ndarray):
img = torch.from_numpy(pic.transpose((2, 0, 1)))
return img
# handle PIL Image
if pic.mode == 'I':
img = torch.from_numpy(np.array(pic, np.int32, copy=False))
elif pic.mode == 'I;16':
img = torch.from_numpy(np.array(pic, np.int16, copy=False))
else:
img = torch.ByteTensor(torch.ByteStorage.from_buffer(pic.tobytes()))
# PIL image mode: 1, L, P, I, F, RGB, YCbCr, RGBA, CMYK
if pic.mode == 'YCbCr':
nchannel = 3
elif pic.mode == 'I;16':
nchannel = 1
else:
nchannel = len(pic.mode)
img = img.view(pic.size[1], pic.size[0], nchannel)
img = img.transpose(0, 1).transpose(0, 2).contiguous()
if isinstance(img, torch.ByteTensor):
return img.float()
else:
return img
class Nyu2DataModule(LightningDataModule):
def __init__(self, config: DictConfig):
super().__init__()
self.config = config
self.train_csv_path = os.path.join(config.DATASET.PATH, config.DATASET.TYPE, 'data' , 'nyu2_train.csv')
self.test_csv_path = os.path.join(config.DATASET.PATH, config.DATASET.TYPE, 'data' , 'nyu2_test.csv')
mode = "train"
self.training_samples = DataLoadPreprocess(self.train_csv_path, self.config, mode, transform=preprocessing_transforms(mode))
self.train_loader = DataLoader(self.training_samples, self.config.SOLVER.BATCHSIZE,
shuffle=True,
num_workers=self.config.SOLVER.NUM_WORKERS,
pin_memory=True,
sampler=None)
mode = "test"
self.testing_samples = DataLoadPreprocess(self.test_csv_path, self.config, mode, transform=preprocessing_transforms(mode))
def train_dataloader(self):
return self.train_loader
def val_dataloader(self, eval=False, shuffle=False):
self.test_loader = DataLoader(self.testing_samples, 1,
shuffle = shuffle,
num_workers=1,
pin_memory=True,
sampler=None)
return self.test_loader
class Nyu2DataModuleTest(LightningDataModule):
def __init__(self, config: DictConfig):
super().__init__()
self.config = config
self.test_csv_path = os.path.join(config.DATASET.PATH, config.DATASET.TYPE, 'test.csv')
mode = "test"
self.testing_samples = DataLoadPreprocess(self.test_csv_path, self.config, mode, transform=preprocessing_transforms(mode))
def val_dataloader(self, eval=False, shuffle=False):
self.test_loader = DataLoader(self.testing_samples, 1,
shuffle = shuffle,
num_workers=1,
pin_memory=True,
sampler=None)
return self.test_loader