-
Notifications
You must be signed in to change notification settings - Fork 0
/
GenerateDatasetSimDirections.py
154 lines (123 loc) · 5.06 KB
/
GenerateDatasetSimDirections.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
from __future__ import print_function, division
import os
import torch
import pandas as pd
from skimage import io, transform
import numpy as np
import matplotlib.pyplot as plt
from torch.utils.data import Dataset, DataLoader
from torchvision import transforms, utils
import imageio
import cv2
import json
from pprint import pprint
# Ignore warnings
import warnings
warnings.filterwarnings("ignore")
plt.ion() # interactive mode
#root_dir = '/media/nigno/Data/simulatedEyesDataset/imgs/'
root_dir = '/media/nigno/Data/UnityProjects/UnityEyes/Build/imgs/'
save_root_dir = '/media/nigno/Data/Projects/EyeNet/Datasets/SimDatasetFull/'
dataset_list = os.listdir(root_dir)
num_images = int(len(dataset_list) / 4)
print (num_images)
raw_input('Press enter to continue: ')
class_num = 30.0
class_factor = (2.0 / class_num)
num_elem_per_dir = 1000
dataset_count = 0
labels = []
labels_dir_left = []
labels_dir_right = []
labels_dist = []
labels_class_left = []
labels_class_right = []
for i in range(num_images):
print(str(i // num_elem_per_dir) + " " + str(dataset_count) )
if (i % num_elem_per_dir) == 0.0:
dataset_count = 0
subset_path = str(i // num_elem_per_dir) + '/'
save_path = save_root_dir + subset_path
if not os.path.isdir(save_path):
os.makedirs(save_path)
# if ((i // num_elem_per_dir) > 0):
# subset_path = str((i // num_elem_per_dir) - 1 ) + '/'
# labels_save_path = save_root_dir + subset_path
# np.save(labels_save_path + 'labels.npy', labels)
# np.save(labels_save_path + 'labels_dir_left.npy', labels_dir_left)
# np.save(labels_save_path + 'labels_dir_right.npy', labels_dir_right)
# np.save(labels_save_path + 'labels_dist.npy', labels_dist)
labels = []
labels_dir_left = []
labels_dir_right = []
labels_dist = []
labels_class_left = []
labels_class_right = []
# if (i % 2) == 0.0:
# json_name = str((i * 2) + 1) + '.json'
#
# with open(root_dir + json_name) as json_data:
# data = json.load(json_data)
# json_data.close()
#
# pitch = float(data['target_3D_pos']['pitch']) * np.pi / 180.0
# yaw = float(data['target_3D_pos']['yaw']) * np.pi / 180.0
# d = float(data['target_3D_pos']['z'])
# x = d * np.cos(pitch) * np.sin(yaw)
# y = d * np.sin(pitch) * np.cos(yaw)
# z = d * np.cos(pitch) * np.cos(yaw)
# labels.append([x, y, z])
json_name = str((i * 2) + 1) + '.json'
with open(root_dir + json_name) as json_data:
data = json.load(json_data)
json_data.close()
x = float(data['target_3D_pos']['x'])
y = float(data['target_3D_pos']['y'])
z = float(data['target_3D_pos']['z'])
x *= (z / 20.0)
y *= (z / 20.0)
labels.append([x, y, z])
dist = float(data['target_3D_pos']['eye_distance'])
labels_dist.append(dist)
x = float(data['eye_3D_vec']['x'])
y = float(data['eye_3D_vec']['y'])
z = float(data['eye_3D_vec']['z'])
labels_dir_left.append([x, y, z])
x_class = (x + 1.0) // class_factor
y_class = (y + 1.0) // class_factor
class_val = int(y_class * class_num + x_class)
labels_class_left.append(class_val)
json_name = str((i * 2) + 2) + '.json'
with open(root_dir + json_name) as json_data:
data = json.load(json_data)
json_data.close()
x = float(data['eye_3D_vec']['x'])
y = float(data['eye_3D_vec']['y'])
z = float(data['eye_3D_vec']['z'])
labels_dir_right.append([x, y, z])
x_class = (x + 1.0) // class_factor
y_class = (y + 1.0) // class_factor
class_val = int(y_class * class_num + x_class)
labels_class_right.append(class_val)
img_name_left = str((i * 2) + 1) + '.jpg'
img_name_right = str((i * 2) + 2) + '.jpg'
cv_left = cv2.imread(root_dir + img_name_left)
cv_right = cv2.imread(root_dir + img_name_right)
image_left = cv2.cvtColor(cv_left, cv2.COLOR_BGR2GRAY)
image_right = cv2.cvtColor(cv_right, cv2.COLOR_BGR2GRAY)
img_name = str(dataset_count) + '.png'
image_path_left = save_path + 'left'
image_path_right = save_path + 'right'
io.imsave(image_path_left + img_name, image_left)
io.imsave(image_path_right + img_name, image_right)
dataset_count += 1
if (((i % num_elem_per_dir) == (num_elem_per_dir - 1)) or (i == (num_images - 1))):
subset_path = str(i // num_elem_per_dir) + '/'
labels_save_path = save_root_dir + subset_path
np.save(labels_save_path + 'labels.npy', labels)
np.save(labels_save_path + 'labels_dir_left.npy', labels_dir_left)
np.save(labels_save_path + 'labels_dir_right.npy', labels_dir_right)
np.save(labels_save_path + 'labels_dist.npy', labels_dist)
np.save(labels_save_path + 'labels_class_left.npy', labels_class_left)
np.save(labels_save_path + 'labels_class_right.npy', labels_class_right)
raw_input('Press enter to continue: ')