ASD-SA/loss_function.py at master · codename1995/ASD-SA · GitHub

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
from __future__ import division
import keras.backend as K

eps=2.2204e-16

# KL-Divergence Loss
def kl_divergence(y_true, y_pred):
    shape_r_out = y_pred._keras_shape[1]
    shape_c_out = y_pred._keras_shape[2]

    min_y_pred = K.repeat_elements(K.expand_dims(
                 K.repeat_elements(K.expand_dims(
                     K.min(K.min(y_pred, axis=2), axis=1)
                     , axis=1), shape_r_out, axis=1)
                     , axis=2), shape_c_out, axis=2)
    y_pred = y_pred - min_y_pred

    max_y_pred = K.repeat_elements(K.expand_dims(
                 K.repeat_elements(K.expand_dims(
                     K.max(K.max(y_pred, axis=2), axis=1)
                     , axis=1), shape_r_out, axis=1)
                     , axis=2), shape_c_out, axis=2)

    y_pred /= max_y_pred

    sum_y_true = K.repeat_elements(K.expand_dims(
                 K.repeat_elements(K.expand_dims(
                     K.sum(K.sum(y_true, axis=2), axis=1)
                     , axis=1), shape_r_out, axis=1)
                     , axis=2), shape_c_out, axis=2)
    sum_y_pred = K.repeat_elements(K.expand_dims(
                 K.repeat_elements(K.expand_dims(
                     K.sum(K.sum(y_pred, axis=2), axis=1)
                     , axis=1), shape_r_out, axis=1)
                     , axis=2), shape_c_out, axis=2)

    y_true /= (sum_y_true + eps)
    y_pred /= (sum_y_pred + eps)

    return K.sum(K.sum(y_true * K.log((y_true / (y_pred + eps)) + eps), axis=2), axis=1)

# Correlation Coefficient Loss
def correlation_coefficient(y_true, y_pred):
    shape_r_out = y_pred._keras_shape[1]
    shape_c_out = y_pred._keras_shape[2]

    min_y_pred = K.repeat_elements(K.expand_dims(
                 K.repeat_elements(K.expand_dims(
                     K.min(K.min(y_pred, axis=2), axis=1)
                     , axis=1), shape_r_out, axis=1)
                     , axis=2), shape_c_out, axis=2)
    y_pred = y_pred - min_y_pred

    max_y_pred = K.repeat_elements(K.expand_dims(
                 K.repeat_elements(K.expand_dims(
                     K.max(K.max(y_pred, axis=2), axis=1)
                     , axis=1), shape_r_out, axis=1)
                     , axis=2), shape_c_out, axis=2)

    y_pred /= max_y_pred

    sum_y_true = K.repeat_elements(K.expand_dims(
                 K.repeat_elements(K.expand_dims(
                     K.sum(K.sum(y_true, axis=2), axis=1)
                     , axis=1), shape_r_out, axis=1)
                     , axis=2), shape_c_out, axis=2)
    sum_y_pred = K.repeat_elements(K.expand_dims(
                 K.repeat_elements(K.expand_dims(
                     K.sum(K.sum(y_pred, axis=2), axis=1)
                     , axis=1), shape_r_out, axis=1)
                     , axis=2), shape_c_out, axis=2)

    y_true /= (sum_y_true + eps)
    y_pred /= (sum_y_pred + eps)

    N = shape_r_out * shape_c_out
    sum_prod = K.sum(K.sum(y_true * y_pred, axis=2), axis=1)
    sum_x = K.sum(K.sum(y_true, axis=2), axis=1)
    sum_y = K.sum(K.sum(y_pred, axis=2), axis=1)
    sum_x_square = K.sum(K.sum(K.square(y_true), axis=2), axis=1)
    sum_y_square = K.sum(K.sum(K.square(y_pred), axis=2), axis=1)

    num = sum_prod - ((sum_x * sum_y) / N)
    den = K.sqrt((sum_x_square - K.square(sum_x) / N) * (sum_y_square - K.square(sum_y) / N))

    return -2 * num / den

# Normalized Scanpath Saliency Loss
def nss(y_true, y_pred):
    shape_r_out = y_pred._keras_shape[1]
    shape_c_out = y_pred._keras_shape[2]

    max_y_pred = K.repeat_elements(K.expand_dims(
                 K.repeat_elements(K.expand_dims(
                     K.max(K.max(y_pred, axis=2), axis=1)
                     , axis=1), shape_r_out, axis=1)
                     , axis=2), shape_c_out, axis=2)

    y_pred /= max_y_pred
    y_pred_flatten = K.batch_flatten(y_pred)

    y_mean = K.mean(y_pred_flatten, axis=-1)
    y_mean = K.repeat_elements(K.expand_dims(
             K.repeat_elements(K.expand_dims(
                 K.expand_dims(y_mean),
                 axis=1), shape_r_out, axis=1),
                 axis=2), shape_c_out, axis=2)

    y_std = K.std(y_pred_flatten, axis=-1)
    y_std = K.repeat_elements(K.expand_dims(
             K.repeat_elements(K.expand_dims(
                 K.expand_dims(y_std),
                 axis=1), shape_r_out, axis=1),
                 axis=2), shape_c_out, axis=2)

    y_pred = (y_pred - y_mean) / (y_std + eps)

    return -(K.sum(K.sum(y_true * y_pred, axis=2), axis=1) / K.sum(K.sum(y_true, axis=2), axis=1))

# Discriminative region enhancement loss
def dre_loss(y_true, y_pred):
    shape_r_out = y_pred._keras_shape[1]
    shape_c_out = y_pred._keras_shape[2]

    # Min-Max Normalization
    min_y_pred = K.repeat_elements(K.expand_dims(
                 K.repeat_elements(K.expand_dims(
                     K.min(K.min(y_pred, axis=2), axis=1)
                     , axis=1), shape_r_out, axis=1)
                     , axis=2), shape_c_out, axis=2)
    y_pred = y_pred - min_y_pred
    max_y_pred = K.repeat_elements(K.expand_dims(
                 K.repeat_elements(K.expand_dims(
                     K.max(K.max(y_pred, axis=2), axis=1)
                     , axis=1), shape_r_out, axis=1)
                     , axis=2), shape_c_out, axis=2)
    y_pred /= (max_y_pred + eps)

    return K.sum(K.sum(y_true * K.abs(y_pred-y_true), axis=2), axis=1) / K.sum(K.sum(y_true, axis=2), axis=1)