python - Keras - 密集层与 Convolution2D 层的融合

Question

我想制作一个自定义层，该层应该将密集层的输出与 Convolution2D 层融合。

创意来自this paper这是网络:

融合层尝试将 Convolution2D 张量 (256x28x28) 与密集张量 (256) 融合。这是它的方程式:

y_global => Dense layer output with shape 256 y_mid => Convolution2D layer output with shape 256x28x28

这是关于融合过程的论文描述:

我最终制作了一个新的自定义图层，如下所示:

class FusionLayer(Layer):

    def __init__(self, output_dim, **kwargs):
        self.output_dim = output_dim
        super(FusionLayer, self).__init__(**kwargs)

    def build(self, input_shape):
        input_dim = input_shape[1][1]
        initial_weight_value = np.random.random((input_dim, self.output_dim))
        self.W = K.variable(initial_weight_value)
        self.b = K.zeros((input_dim,))
        self.trainable_weights = [self.W, self.b]

    def call(self, inputs, mask=None):
        y_global = inputs[0]
        y_mid = inputs[1]
        # the code below should be modified
        output = K.dot(K.concatenate([y_global, y_mid]), self.W)
        output += self.b
        return self.activation(output)

    def get_output_shape_for(self, input_shape):
        assert input_shape and len(input_shape) == 2
        return (input_shape[0], self.output_dim)

我想我得到了 __init__和 build方法正确，但我不知道如何连接 y_global (256 个维度)与 y-mid (256x28x28 尺寸)在 call层，以便输出与上述等式相同。

如何在 call 中实现这个等式方法？

非常感谢...

更新:任何其他成功整合这两层数据的方法对我来说也是可以接受的……它不一定非得像论文中提到的那样，但它至少需要返回一个可接受的输出。 ..

Answer 1

我在做一个图像着色项目，最后遇到了一个融合层问题，然后我找到了一个包含融合层的模型。 Hope 在一定程度上解决了您的问题。

    embed_input = Input(shape=(1000,))
    encoder_input = Input(shape=(256, 256, 1,))

    #Encoder
    encoder_output = Conv2D(64, (3,3), activation='relu', padding='same', strides=2,
                            bias_initializer=TruncatedNormal(mean=0.0, stddev=0.05))(encoder_input)
    encoder_output = Conv2D(128, (3,3), activation='relu', padding='same',
                            bias_initializer=TruncatedNormal(mean=0.0, stddev=0.05))(encoder_output)
    encoder_output = Conv2D(128, (3,3), activation='relu', padding='same', strides=2,
                            bias_initializer=TruncatedNormal(mean=0.0, stddev=0.05))(encoder_output)
    encoder_output = Conv2D(256, (3,3), activation='relu', padding='same',
                            bias_initializer=TruncatedNormal(mean=0.0, stddev=0.05))(encoder_output)
    encoder_output = Conv2D(256, (3,3), activation='relu', padding='same', strides=2,
                            bias_initializer=TruncatedNormal(mean=0.0, stddev=0.05))(encoder_output)
    encoder_output = Conv2D(512, (3,3), activation='relu', padding='same',
                            bias_initializer=TruncatedNormal(mean=0.0, stddev=0.05))(encoder_output)
    encoder_output = Conv2D(512, (3,3), activation='relu', padding='same',
                            bias_initializer=TruncatedNormal(mean=0.0, stddev=0.05))(encoder_output)
    encoder_output = Conv2D(256, (3,3), activation='relu', padding='same',
                            bias_initializer=TruncatedNormal(mean=0.0, stddev=0.05))(encoder_output)

    #Fusion
    fusion_output = RepeatVector(32 * 32)(embed_input)
    fusion_output = Reshape(([32, 32, 1000]))(fusion_output)
    fusion_output = concatenate([encoder_output, fusion_output], axis=3)
    fusion_output = Conv2D(256, (1, 1), activation='relu', padding='same',
                            bias_initializer=TruncatedNormal(mean=0.0, stddev=0.05))(fusion_output)

    #Decoder
    decoder_output = Conv2D(128, (3,3), activation='relu', padding='same',
                            bias_initializer=TruncatedNormal(mean=0.0, stddev=0.05))(fusion_output)
    decoder_output = UpSampling2D((2, 2))(decoder_output)
    decoder_output = Conv2D(64, (3,3), activation='relu', padding='same',
                            bias_initializer=TruncatedNormal(mean=0.0, stddev=0.05))(decoder_output)
    decoder_output = UpSampling2D((2, 2))(decoder_output)
    decoder_output = Conv2D(32, (3,3), activation='relu', padding='same',
                            bias_initializer=TruncatedNormal(mean=0.0, stddev=0.05))(decoder_output)
    decoder_output = Conv2D(16, (3,3), activation='relu', padding='same',
                            bias_initializer=TruncatedNormal(mean=0.0, stddev=0.05))(decoder_output)
    decoder_output = Conv2D(2, (3, 3), activation='tanh', padding='same',
                            bias_initializer=TruncatedNormal(mean=0.0, stddev=0.05))(decoder_output)
    decoder_output = UpSampling2D((2, 2))(decoder_output)

    model = Model(inputs=[encoder_input, embed_input], outputs=decoder_output)

这是源链接: https://github.com/hvvashistha/Auto-Colorize