-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathdataloader.py
148 lines (124 loc) · 5.48 KB
/
dataloader.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
from torch.utils.data import Dataset
import torch as th
from PIL import Image
import pickle
import numpy as np
import cv2
class Surgical_dataset(Dataset):
def __init__(self, data_dir):
with open(data_dir, "rb") as f:
self.img_list = pickle.load(f)
def __len__(self):
return len(self.img_list)
def __getitem__(self, idx):
direct = self.img_list[idx]
path = direct["path"]
left_t0 = path + "img_left/" + direct["t0"] + ".jpg"
left_t0 = np.array(Image.open(left_t0))
left_t1 = path + "img_left/" + direct["t1"] + ".jpg"
left_t1 = np.array(Image.open(left_t1))
left_t2 = path + "img_left/" + direct["t2"] + ".jpg"
left_t2 = np.array(Image.open(left_t2))
left_t3 = path + "img_left/" + direct["t3"] + ".jpg"
left_t3 = np.array(Image.open(left_t3))
left_t4 = path + "img_left/" + direct["t4"] + ".jpg"
left_t4 = np.array(Image.open(left_t4))
right_t0 = path + "img_right/" + direct["t0"] + ".jpg"
right_t0 = np.array(Image.open(right_t0))
right_t1 = path + "img_right/" + direct["t1"] + ".jpg"
right_t1 = np.array(Image.open(right_t1))
right_t2 = path + "img_right/" + direct["t2"] + ".jpg"
right_t2 = np.array(Image.open(right_t2))
right_t3 = path + "img_right/" + direct["t3"] + ".jpg"
right_t3 = np.array(Image.open(right_t3))
right_t4 = path + "img_right/" + direct["t4"] + ".jpg"
right_t4 = np.array(Image.open(right_t4))
disp_t0 = path + "disp/" + direct["t0"] + ".npy"
disp_t0 = np.load(disp_t0)
disp_t1 = path + "disp/" + direct["t1"] + ".npy"
disp_t1 = np.load(disp_t1)
disp_t2 = path + "disp/" + direct["t2"] + ".npy"
disp_t2 = np.load(disp_t2)
disp_t3 = path + "disp/" + direct["t3"] + ".npy"
disp_t3 = np.load(disp_t3)
disp_t4 = path + "disp/" + direct["t4"] + ".npy"
disp_t4 = np.load(disp_t4)
tool_t0 = path + "tool_mask/" + direct["t0"] + ".png"
tool_t0 = cv2.imread(tool_t0, -1)
tool_t0 = tool_t0[:512, 226:-2]
tool_t1 = path + "tool_mask/" + direct["t1"] + ".png"
tool_t1 = cv2.imread(tool_t1, -1)
tool_t1 = tool_t1[:512, 226:-2]
tool_t2 = path + "tool_mask/" + direct["t2"] + ".png"
tool_t2 = cv2.imread(tool_t2, -1)
tool_t2 = tool_t2[:512, 226:-2]
tool_t3 = path + "tool_mask/" + direct["t3"] + ".png"
tool_t3 = cv2.imread(tool_t3, -1)
tool_t3 = tool_t3[:512, 226:-2]
tool_t4 = path + "tool_mask/" + direct["t4"] + ".png"
tool_t4 = cv2.imread(tool_t4, -1)
tool_t4 = tool_t4[:512, 226:-2]
tool_t0_r = path + "tool_mask_right/" + direct["t0"] + ".png"
tool_t0_r = cv2.imread(tool_t0_r, -1)
tool_t1_r = path + "tool_mask_right/" + direct["t1"] + ".png"
tool_t1_r = cv2.imread(tool_t1_r, -1)
tool_t2_r = path + "tool_mask_right/" + direct["t2"] + ".png"
tool_t2_r = cv2.imread(tool_t2_r, -1)
tool_t3_r = path + "tool_mask_right/" + direct["t3"] + ".png"
tool_t3_r = cv2.imread(tool_t3_r, -1)
tool_t4_r = path + "tool_mask_right/" + direct["t4"] + ".png"
tool_t4_r = cv2.imread(tool_t4_r, -1)
res = {}
for i in ["left_t0", "left_t1", "left_t2", "left_t3", "left_t4"]:
val = vars()[i]
val = val[:512, 226:-2, :]
val = th.from_numpy(val.astype(np.float32)).permute(2, 0, 1) / 255.
res[i] = val
for i in ["right_t0", "right_t1", "right_t2", "right_t3", "right_t4"]:
val = vars()[i]
val = th.from_numpy(val.astype(np.float32)).permute(2, 0, 1) / 255.
res[i] = val
for i in ["disp_t0", "disp_t1", "disp_t2", "disp_t3", "disp_t4"]:
val = vars()[i]
val = val[:512, 226:-2]
val = th.from_numpy(val.astype(np.float32)).unsqueeze(0)
res[i] = val
for i in ["tool_t0", "tool_t1", "tool_t2", "tool_t3", "tool_t4",
"tool_t0_r", "tool_t1_r", "tool_t2_r", "tool_t3_r", "tool_t4_r"]:
val = vars()[i]
val = th.from_numpy(val.astype(np.float32)).unsqueeze(0)
res[i] = val
return res
class Surgical_dataset_eval(Dataset):
def __init__(self, data_dir):
with open(data_dir, "rb") as f:
self.img_list = pickle.load(f)
def __len__(self):
return len(self.img_list)
def __getitem__(self, idx):
direct = self.img_list[idx]
path = direct["path"]
left = []
disp = []
tool = []
for i in direct['sequence']:
left_t0 = path + "img_left/" + i + ".jpg"
left_t0 = np.array(Image.open(left_t0))
left_t0 = left_t0[:512, 226:-2, :]
left_t0 = th.from_numpy(left_t0.astype(np.float32)).permute(2, 0, 1) / 255.
left.append(left_t0)
disp_t0 = path + "disp/" + i + ".npy"
disp_t0 = np.load(disp_t0)
disp_t0 = disp_t0[:512, 226:-2]
disp_t0 = th.from_numpy(disp_t0.astype(np.float32)).unsqueeze(0)
disp.append(disp_t0)
tool_t0 = path + "tool_mask/" + i + ".png"
tool_t0 = cv2.imread(tool_t0, -1)
tool_t0 = tool_t0[:512, 226:-2]
tool_t0 = th.from_numpy(tool_t0.astype(np.float32)).unsqueeze(0)
tool.append(tool_t0)
res = {}
res["left"] = left
res["disp"] = disp
res["tool"] = tool
return res