File size: 7,024 Bytes
d061c3e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import torch
import torch.nn.functional as F
from math import exp
import numpy as np

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")


def gaussian(window_size, sigma):
    gauss = torch.Tensor([exp(-((x - window_size // 2) ** 2) / float(2 * sigma**2)) for x in range(window_size)])
    return gauss / gauss.sum()


def create_window(window_size, channel=1):
    _1D_window = gaussian(window_size, 1.5).unsqueeze(1)
    _2D_window = _1D_window.mm(_1D_window.t()).float().unsqueeze(0).unsqueeze(0).to(device)
    window = _2D_window.expand(channel, 1, window_size, window_size).contiguous()
    return window


def create_window_3d(window_size, channel=1):
    _1D_window = gaussian(window_size, 1.5).unsqueeze(1)
    _2D_window = _1D_window.mm(_1D_window.t())
    _3D_window = _2D_window.unsqueeze(2) @ (_1D_window.t())
    window = _3D_window.expand(1, channel, window_size, window_size, window_size).contiguous().to(device)
    return window


def ssim(img1, img2, window_size=11, window=None, size_average=True, full=False, val_range=None):
    # Value range can be different from 255. Other common ranges are 1 (sigmoid) and 2 (tanh).
    if val_range is None:
        if torch.max(img1) > 128:
            max_val = 255
        else:
            max_val = 1

        if torch.min(img1) < -0.5:
            min_val = -1
        else:
            min_val = 0
        L = max_val - min_val
    else:
        L = val_range

    padd = 0
    (_, channel, height, width) = img1.size()
    if window is None:
        real_size = min(window_size, height, width)
        window = create_window(real_size, channel=channel).to(img1.device)

    # mu1 = F.conv2d(img1, window, padding=padd, groups=channel)
    # mu2 = F.conv2d(img2, window, padding=padd, groups=channel)
    mu1 = F.conv2d(F.pad(img1, (5, 5, 5, 5), mode="replicate"), window, padding=padd, groups=channel)
    mu2 = F.conv2d(F.pad(img2, (5, 5, 5, 5), mode="replicate"), window, padding=padd, groups=channel)

    mu1_sq = mu1.pow(2)
    mu2_sq = mu2.pow(2)
    mu1_mu2 = mu1 * mu2

    sigma1_sq = F.conv2d(F.pad(img1 * img1, (5, 5, 5, 5), "replicate"), window, padding=padd, groups=channel) - mu1_sq
    sigma2_sq = F.conv2d(F.pad(img2 * img2, (5, 5, 5, 5), "replicate"), window, padding=padd, groups=channel) - mu2_sq
    sigma12 = F.conv2d(F.pad(img1 * img2, (5, 5, 5, 5), "replicate"), window, padding=padd, groups=channel) - mu1_mu2

    C1 = (0.01 * L) ** 2
    C2 = (0.03 * L) ** 2

    v1 = 2.0 * sigma12 + C2
    v2 = sigma1_sq + sigma2_sq + C2
    cs = torch.mean(v1 / v2)  # contrast sensitivity

    ssim_map = ((2 * mu1_mu2 + C1) * v1) / ((mu1_sq + mu2_sq + C1) * v2)

    if size_average:
        ret = ssim_map.mean()
    else:
        ret = ssim_map.mean(1).mean(1).mean(1)

    if full:
        return ret, cs
    return ret


def ssim_matlab(img1, img2, window_size=11, window=None, size_average=True, full=False, val_range=None):
    # Value range can be different from 255. Other common ranges are 1 (sigmoid) and 2 (tanh).
    if val_range is None:
        if torch.max(img1) > 128:
            max_val = 255
        else:
            max_val = 1

        if torch.min(img1) < -0.5:
            min_val = -1
        else:
            min_val = 0
        L = max_val - min_val
    else:
        L = val_range

    padd = 0
    (_, _, height, width) = img1.size()
    if window is None:
        real_size = min(window_size, height, width)
        window = create_window_3d(real_size, channel=1).to(img1.device, dtype=img1.dtype)
        # Channel is set to 1 since we consider color images as volumetric images

    img1 = img1.unsqueeze(1)
    img2 = img2.unsqueeze(1)

    mu1 = F.conv3d(F.pad(img1, (5, 5, 5, 5, 5, 5), mode="replicate"), window, padding=padd, groups=1)
    mu2 = F.conv3d(F.pad(img2, (5, 5, 5, 5, 5, 5), mode="replicate"), window, padding=padd, groups=1)

    mu1_sq = mu1.pow(2)
    mu2_sq = mu2.pow(2)
    mu1_mu2 = mu1 * mu2

    sigma1_sq = F.conv3d(F.pad(img1 * img1, (5, 5, 5, 5, 5, 5), "replicate"), window, padding=padd, groups=1) - mu1_sq
    sigma2_sq = F.conv3d(F.pad(img2 * img2, (5, 5, 5, 5, 5, 5), "replicate"), window, padding=padd, groups=1) - mu2_sq
    sigma12 = F.conv3d(F.pad(img1 * img2, (5, 5, 5, 5, 5, 5), "replicate"), window, padding=padd, groups=1) - mu1_mu2

    C1 = (0.01 * L) ** 2
    C2 = (0.03 * L) ** 2

    v1 = 2.0 * sigma12 + C2
    v2 = sigma1_sq + sigma2_sq + C2
    cs = torch.mean(v1 / v2)  # contrast sensitivity

    ssim_map = ((2 * mu1_mu2 + C1) * v1) / ((mu1_sq + mu2_sq + C1) * v2)

    if size_average:
        ret = ssim_map.mean()
    else:
        ret = ssim_map.mean(1).mean(1).mean(1)

    if full:
        return ret, cs
    return ret


def msssim(img1, img2, window_size=11, size_average=True, val_range=None, normalize=False):
    device = img1.device
    weights = torch.FloatTensor([0.0448, 0.2856, 0.3001, 0.2363, 0.1333]).to(device)
    levels = weights.size()[0]
    mssim = []
    mcs = []
    for _ in range(levels):
        sim, cs = ssim(img1, img2, window_size=window_size, size_average=size_average, full=True, val_range=val_range)
        mssim.append(sim)
        mcs.append(cs)

        img1 = F.avg_pool2d(img1, (2, 2))
        img2 = F.avg_pool2d(img2, (2, 2))

    mssim = torch.stack(mssim)
    mcs = torch.stack(mcs)

    # Normalize (to avoid NaNs during training unstable models, not compliant with original definition)
    if normalize:
        mssim = (mssim + 1) / 2
        mcs = (mcs + 1) / 2

    pow1 = mcs**weights
    pow2 = mssim**weights
    # From Matlab implementation https://ece.uwaterloo.ca/~z70wang/research/iwssim/
    output = torch.prod(pow1[:-1] * pow2[-1])
    return output


# Classes to re-use window
class SSIM(torch.nn.Module):
    def __init__(self, window_size=11, size_average=True, val_range=None):
        super(SSIM, self).__init__()
        self.window_size = window_size
        self.size_average = size_average
        self.val_range = val_range

        # Assume 3 channel for SSIM
        self.channel = 3
        self.window = create_window(window_size, channel=self.channel)

    def forward(self, img1, img2):
        (_, channel, _, _) = img1.size()

        if channel == self.channel and self.window.dtype == img1.dtype:
            window = self.window
        else:
            window = create_window(self.window_size, channel).to(img1.device).type(img1.dtype)
            self.window = window
            self.channel = channel

        _ssim = ssim(img1, img2, window=window, window_size=self.window_size, size_average=self.size_average)
        dssim = (1 - _ssim) / 2
        return dssim


class MSSSIM(torch.nn.Module):
    def __init__(self, window_size=11, size_average=True, channel=3):
        super(MSSSIM, self).__init__()
        self.window_size = window_size
        self.size_average = size_average
        self.channel = channel

    def forward(self, img1, img2):
        return msssim(img1, img2, window_size=self.window_size, size_average=self.size_average)