def inference(image, keep_prob):
    """
    Semantic segmentation network definition
    :param image: input image. Should have values in range 0-255
    :param keep_prob:
    :return:
    """
    print(">> Setting up resnet-101 pretrained layers ...")

    resnet101_model = utils.get_model_data(FLAGS.model_dir)
    weights = np.squeeze(resnet101_model['params'])

    mean_pixel = resnet101_model['meta'][0][0][2][0][0][2]
    normalised_img = utils.process_image(image, mean_pixel)

    with tf.variable_scope("inference"):
        net = resnet101_net(normalised_img, weights, keep_prob)
        last_layer = net["res5c_relu"]

        fc_filter = utils.weight_variable([1, 1, 2048, NUM_OF_CLASSES],
                                          name="fc_filter")
        fc_bias = utils.bias_variable([NUM_OF_CLASSES], name="fc_bias")
        fc = tf.nn.bias_add(tf.nn.conv2d(last_layer,
                                         fc_filter,
                                         strides=[1, 1, 1, 1],
                                         padding="SAME"),
                            fc_bias,
                            name='fc')

        # now to upscale to actual image size
        deconv_shape1 = net["res4b22_relu"].get_shape()
        W_t1 = utils.weight_variable(
            [4, 4, deconv_shape1[3].value, NUM_OF_CLASSES], name="W_t1")
        b_t1 = utils.bias_variable([deconv_shape1[3].value], name="b_t1")
        conv_t1 = utils.conv2d_transpose_strided(fc,
                                                 W_t1,
                                                 b_t1,
                                                 output_shape=tf.shape(
                                                     net["res4b22_relu"]))
        fuse_1 = tf.add(conv_t1, net["res4b22_relu"], name="fuse_1")

        deconv_shape2 = net["res3b3_relu"].get_shape()
        W_t2 = utils.weight_variable(
            [4, 4, deconv_shape2[3].value, deconv_shape1[3].value],
            name="W_t2")
        b_t2 = utils.bias_variable([deconv_shape2[3].value], name="b_t2")
        conv_t2 = utils.conv2d_transpose_strided(fuse_1,
                                                 W_t2,
                                                 b_t2,
                                                 output_shape=tf.shape(
                                                     net["res3b3_relu"]))
        fuse_2 = tf.add(conv_t2, net["res3b3_relu"], name="fuse_2")

        shape = tf.shape(image)
        deconv_shape3 = tf.stack(
            [shape[0], shape[1], shape[2], NUM_OF_CLASSES])
        W_t3 = utils.weight_variable(
            [16, 16, NUM_OF_CLASSES, deconv_shape2[3].value], name="W_t3")
        b_t3 = utils.bias_variable([NUM_OF_CLASSES], name="b_t3")
        conv_t3 = utils.conv2d_transpose_strided(fuse_2,
                                                 W_t3,
                                                 b_t3,
                                                 output_shape=deconv_shape3,
                                                 stride=8)

        annotation_pred = tf.argmax(conv_t3, axis=3, name="prediction")

    return tf.expand_dims(annotation_pred, dim=3), conv_t3
示例#2
0
def inference(image, keep_prob):
    """
    Semantic segmentation network definition
    :param image: input image. Should have values in range 0-255
    :param keep_prob:
    :return:
    """
    print("setting up vgg pretrained conv layers ...")
    model_data = utils.get_model_data(model_dir)

    mean = model_data['normalization'][0][0][0]
    mean_pixel = np.mean(mean, axis=(0, 1))
    weights = np.squeeze(model_data['layers'])

    processed_image = utils.process_image(image, mean_pixel)

    with tf.variable_scope("inference"):
        image_net = vgg_net(weights, processed_image)
        conv_final_layer = image_net["conv5_3"]

        pool5 = utils.max_pool_2x2(conv_final_layer)

        W6 = utils.weight_variable([7, 7, 512, 4096], name="W6")
        b6 = utils.bias_variable([4096], name="b6")
        conv6 = utils.conv2d_basic(pool5, W6, b6)
        relu6 = tf.nn.relu(conv6, name="relu6")
        # if FLAGS.debug:
        #     utils.add_activation_summary(relu6)
        relu_dropout6 = tf.nn.dropout(relu6, keep_prob=keep_prob)

        W7 = utils.weight_variable([1, 1, 4096, 4096], name="W7")
        b7 = utils.bias_variable([4096], name="b7")
        conv7 = utils.conv2d_basic(relu_dropout6, W7, b7)
        relu7 = tf.nn.relu(conv7, name="relu7")
        # if FLAGS.debug:
        #     utils.add_activation_summary(relu7)
        relu_dropout7 = tf.nn.dropout(relu7, keep_prob=keep_prob)

        W8 = utils.weight_variable([1, 1, 4096, NUM_OF_CLASSES], name="W8")
        b8 = utils.bias_variable([NUM_OF_CLASSES], name="b8")
        conv8 = utils.conv2d_basic(relu_dropout7, W8, b8)
        annotation_pred1 = tf.argmax(conv8, axis=3, name="prediction1")

        # now to upscale to actual image size
        deconv_shape1 = image_net["pool4"].get_shape()
        W_t1 = utils.weight_variable(
            [4, 4, deconv_shape1[3].value, NUM_OF_CLASSES], name="W_t1")
        b_t1 = utils.bias_variable([deconv_shape1[3].value], name="b_t1")
        conv_t1 = utils.conv2d_transpose_strided(conv8,
                                                 W_t1,
                                                 b_t1,
                                                 output_shape=tf.shape(
                                                     image_net["pool4"]))
        fuse_1 = tf.add(conv_t1, image_net["pool4"], name="fuse_1")

        deconv_shape2 = image_net["pool3"].get_shape()
        W_t2 = utils.weight_variable(
            [4, 4, deconv_shape2[3].value, deconv_shape1[3].value],
            name="W_t2")
        b_t2 = utils.bias_variable([deconv_shape2[3].value], name="b_t2")
        conv_t2 = utils.conv2d_transpose_strided(fuse_1,
                                                 W_t2,
                                                 b_t2,
                                                 output_shape=tf.shape(
                                                     image_net["pool3"]))
        fuse_2 = tf.add(conv_t2, image_net["pool3"], name="fuse_2")

        shape = tf.shape(image)
        deconv_shape3 = tf.stack(
            [shape[0], shape[1], shape[2], NUM_OF_CLASSES])
        W_t3 = utils.weight_variable(
            [16, 16, NUM_OF_CLASSES, deconv_shape2[3].value], name="W_t3")
        b_t3 = utils.bias_variable([NUM_OF_CLASSES], name="b_t3")
        conv_t3 = utils.conv2d_transpose_strided(fuse_2,
                                                 W_t3,
                                                 b_t3,
                                                 output_shape=deconv_shape3,
                                                 stride=8)

        annotation_pred = tf.argmax(conv_t3, axis=3, name="prediction")

    return tf.expand_dims(annotation_pred, dim=3), conv_t3