Training for hopenet works.

natanielruiz

2017-07-07 6664c6d52fad58e396861946a3bed7d5afc4d44d

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
import numpy as np
import torch
import cv2
from torch.utils.data.dataset import Dataset
import os
from PIL import Image
 
import utils
 
class Pose_300W_LP(Dataset):
    def __init__(self, data_dir, filename_path, transform, img_ext='.jpg', annot_ext='.mat'):
        self.data_dir = data_dir
        self.transform = transform
        self.img_ext = img_ext
        self.annot_ext = annot_ext
 
        filename_list = get_list_from_filenames(filename_path)
 
        self.X_train = filename_list
        self.y_train = filename_list
        self.length = len(filename_list)
 
    def __getitem__(self, index):
        img = Image.open(os.path.join(self.data_dir, self.X_train[index] + self.img_ext))
        img = img.convert('RGB')
 
        pose = utils.get_ypr_from_mat(os.path.join(self.data_dir, self.y_train[index] + self.annot_ext))
        label = torch.FloatTensor(pose)
 
        if self.transform is not None:
            img = self.transform(img)
 
        return img, label, self.X_train[index]
 
    def __len__(self):
        # 122,450
        return self.length
 
class AFLW2000(Dataset):
    def __init__(self, data_dir, filename_path, transform, img_ext='.jpg', annot_ext='.mat'):
        self.data_dir = data_dir
        self.transform = transform
        self.img_ext = img_ext
        self.annot_ext = annot_ext
 
        filename_list = get_list_from_filenames(filename_path)
 
        self.X_train = filename_list
        self.y_train = filename_list
        self.length = len(filename_list)
 
    def __getitem__(self, index):
        img = Image.open(os.path.join(self.data_dir, self.X_train[index] + self.img_ext))
        img = img.convert('RGB')
 
        pose = utils.get_ypr_from_mat(os.path.join(self.data_dir, self.y_train[index] + self.annot_ext))
        label = torch.FloatTensor(pose)
 
        if self.transform is not None:
            img = self.transform(img)
 
        return img, label, self.X_train[index]
 
    def __len__(self):
        # 2,000
        return self.length
 
class Pose_300W_LP_binned(Dataset):
    def __init__(self, data_dir, filename_path, transform, img_ext='.jpg', annot_ext='.mat'):
        self.data_dir = data_dir
        self.transform = transform
        self.img_ext = img_ext
        self.annot_ext = annot_ext
 
        filename_list = get_list_from_filenames(filename_path)
 
        self.X_train = filename_list
        self.y_train = filename_list
        self.length = len(filename_list)
 
    def __getitem__(self, index):
        img = Image.open(os.path.join(self.data_dir, self.X_train[index] + self.img_ext))
        img = img.convert('RGB')
 
        # We get the pose in radians
        pose = utils.get_ypr_from_mat(os.path.join(self.data_dir, self.y_train[index] + self.annot_ext))
        # And convert to degrees.
        pitch = pose[0] * 180 / np.pi
        yaw = pose[1] * 180 / np.pi
        roll = pose[2] * 180 / np.pi
        # Bin values
        bins = np.array(range(-99, 102, 3))
        labels = torch.LongTensor(np.digitize([yaw, pitch, roll], bins) - 1)
 
        if self.transform is not None:
            img = self.transform(img)
 
        return img, labels, self.X_train[index]
 
    def __len__(self):
        # 122,450
        return self.length
 
class AFLW2000_binned(Dataset):
    def __init__(self, data_dir, filename_path, transform, img_ext='.jpg', annot_ext='.mat'):
        self.data_dir = data_dir
        self.transform = transform
        self.img_ext = img_ext
        self.annot_ext = annot_ext
 
        filename_list = get_list_from_filenames(filename_path)
 
        self.X_train = filename_list
        self.y_train = filename_list
        self.length = len(filename_list)
 
    def __getitem__(self, index):
        img = Image.open(os.path.join(self.data_dir, self.X_train[index] + self.img_ext))
        img = img.convert('RGB')
 
        # We get the pose in radians
        pose = utils.get_ypr_from_mat(os.path.join(self.data_dir, self.y_train[index] + self.annot_ext))
        # And convert to degrees.
        pitch, yaw, roll = pose * 180 / np.pi
        # Bin values
        bins = np.array(range(-99, 102, 3))
        binned_pitch = torch.DoubleTensor(np.digitize(pitch, bins) - 1)
        binned_yaw = torch.DoubleTensor(np.digitize(yaw, bins) - 1)
        binned_roll = torch.DoubleTensor(np.digitize(roll, bins) - 1)
 
        label = binned_yaw, binned_pitch, binned_roll
 
        if self.transform is not None:
            img = self.transform(img)
 
        return img, label, self.X_train[index]
 
    def __len__(self):
        # 2,000
        return self.length
 
def get_list_from_filenames(file_path):
    # input:    relative path to .txt file with file names
    # output:   list of relative path names
    with open(file_path) as f:
        lines = f.read().splitlines()
    return lines