PaddleX/dygraph/paddlex/paddleseg/models/u2net.py

# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#    http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

import paddle
import paddle.nn as nn
import paddle.nn.functional as F

from paddlex.paddleseg.cvlibs import manager
from paddlex.paddleseg.models import layers
from paddlex.paddleseg.utils import utils

__all__ = ['U2Net', 'U2Netp']


@manager.MODELS.add_component
class U2Net(nn.Layer):
    """
    The U^2-Net implementation based on PaddlePaddle.

    The original article refers to
    Xuebin Qin, et, al. "U^2-Net: Going Deeper with Nested U-Structure for Salient Object Detection"
    (https://arxiv.org/abs/2005.09007).

    Args:
        num_classes (int): The unique number of target classes.
        in_ch (int, optional): Input channels. Default: 3.
        pretrained (str, optional): The path or url of pretrained model for fine tuning. Default: None.

    """

    def __init__(self, num_classes, in_ch=3, pretrained=None):
        super(U2Net, self).__init__()

        self.stage1 = RSU7(in_ch, 32, 64)
        self.pool12 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.stage2 = RSU6(64, 32, 128)
        self.pool23 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.stage3 = RSU5(128, 64, 256)
        self.pool34 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.stage4 = RSU4(256, 128, 512)
        self.pool45 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.stage5 = RSU4F(512, 256, 512)
        self.pool56 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.stage6 = RSU4F(512, 256, 512)

        # decoder
        self.stage5d = RSU4F(1024, 256, 512)
        self.stage4d = RSU4(1024, 128, 256)
        self.stage3d = RSU5(512, 64, 128)
        self.stage2d = RSU6(256, 32, 64)
        self.stage1d = RSU7(128, 16, 64)

        self.side1 = nn.Conv2D(64, num_classes, 3, padding=1)
        self.side2 = nn.Conv2D(64, num_classes, 3, padding=1)
        self.side3 = nn.Conv2D(128, num_classes, 3, padding=1)
        self.side4 = nn.Conv2D(256, num_classes, 3, padding=1)
        self.side5 = nn.Conv2D(512, num_classes, 3, padding=1)
        self.side6 = nn.Conv2D(512, num_classes, 3, padding=1)

        self.outconv = nn.Conv2D(6 * num_classes, num_classes, 1)

        self.pretrained = pretrained
        self.init_weight()

    def forward(self, x):

        hx = x

        #stage 1
        hx1 = self.stage1(hx)
        hx = self.pool12(hx1)

        #stage 2
        hx2 = self.stage2(hx)
        hx = self.pool23(hx2)

        #stage 3
        hx3 = self.stage3(hx)
        hx = self.pool34(hx3)

        #stage 4
        hx4 = self.stage4(hx)
        hx = self.pool45(hx4)

        #stage 5
        hx5 = self.stage5(hx)
        hx = self.pool56(hx5)

        #stage 6
        hx6 = self.stage6(hx)
        hx6up = _upsample_like(hx6, hx5)

        #-------------------- decoder --------------------
        hx5d = self.stage5d(paddle.concat((hx6up, hx5), 1))
        hx5dup = _upsample_like(hx5d, hx4)

        hx4d = self.stage4d(paddle.concat((hx5dup, hx4), 1))
        hx4dup = _upsample_like(hx4d, hx3)

        hx3d = self.stage3d(paddle.concat((hx4dup, hx3), 1))
        hx3dup = _upsample_like(hx3d, hx2)

        hx2d = self.stage2d(paddle.concat((hx3dup, hx2), 1))
        hx2dup = _upsample_like(hx2d, hx1)

        hx1d = self.stage1d(paddle.concat((hx2dup, hx1), 1))

        #side output
        d1 = self.side1(hx1d)

        d2 = self.side2(hx2d)
        d2 = _upsample_like(d2, d1)

        d3 = self.side3(hx3d)
        d3 = _upsample_like(d3, d1)

        d4 = self.side4(hx4d)
        d4 = _upsample_like(d4, d1)

        d5 = self.side5(hx5d)
        d5 = _upsample_like(d5, d1)

        d6 = self.side6(hx6)
        d6 = _upsample_like(d6, d1)

        d0 = self.outconv(paddle.concat((d1, d2, d3, d4, d5, d6), 1))

        return [d0, d1, d2, d3, d4, d5, d6]

    def init_weight(self):
        if self.pretrained is not None:
            utils.load_entire_model(self, self.pretrained)


### U^2-Net small ###
@manager.MODELS.add_component
class U2Netp(nn.Layer):
    """Please Refer to U2Net above."""

    def __init__(self, num_classes, in_ch=3, pretrained=None):
        super(U2Netp, self).__init__()

        self.stage1 = RSU7(in_ch, 16, 64)
        self.pool12 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.stage2 = RSU6(64, 16, 64)
        self.pool23 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.stage3 = RSU5(64, 16, 64)
        self.pool34 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.stage4 = RSU4(64, 16, 64)
        self.pool45 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.stage5 = RSU4F(64, 16, 64)
        self.pool56 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.stage6 = RSU4F(64, 16, 64)

        # decoder
        self.stage5d = RSU4F(128, 16, 64)
        self.stage4d = RSU4(128, 16, 64)
        self.stage3d = RSU5(128, 16, 64)
        self.stage2d = RSU6(128, 16, 64)
        self.stage1d = RSU7(128, 16, 64)

        self.side1 = nn.Conv2D(64, num_classes, 3, padding=1)
        self.side2 = nn.Conv2D(64, num_classes, 3, padding=1)
        self.side3 = nn.Conv2D(64, num_classes, 3, padding=1)
        self.side4 = nn.Conv2D(64, num_classes, 3, padding=1)
        self.side5 = nn.Conv2D(64, num_classes, 3, padding=1)
        self.side6 = nn.Conv2D(64, num_classes, 3, padding=1)

        self.outconv = nn.Conv2D(6 * num_classes, num_classes, 1)

        self.pretrained = pretrained
        self.init_weight()

    def forward(self, x):

        hx = x

        #stage 1
        hx1 = self.stage1(hx)
        hx = self.pool12(hx1)

        #stage 2
        hx2 = self.stage2(hx)
        hx = self.pool23(hx2)

        #stage 3
        hx3 = self.stage3(hx)
        hx = self.pool34(hx3)

        #stage 4
        hx4 = self.stage4(hx)
        hx = self.pool45(hx4)

        #stage 5
        hx5 = self.stage5(hx)
        hx = self.pool56(hx5)

        #stage 6
        hx6 = self.stage6(hx)
        hx6up = _upsample_like(hx6, hx5)

        #decoder
        hx5d = self.stage5d(paddle.concat((hx6up, hx5), 1))
        hx5dup = _upsample_like(hx5d, hx4)

        hx4d = self.stage4d(paddle.concat((hx5dup, hx4), 1))
        hx4dup = _upsample_like(hx4d, hx3)

        hx3d = self.stage3d(paddle.concat((hx4dup, hx3), 1))
        hx3dup = _upsample_like(hx3d, hx2)

        hx2d = self.stage2d(paddle.concat((hx3dup, hx2), 1))
        hx2dup = _upsample_like(hx2d, hx1)

        hx1d = self.stage1d(paddle.concat((hx2dup, hx1), 1))

        #side output
        d1 = self.side1(hx1d)

        d2 = self.side2(hx2d)
        d2 = _upsample_like(d2, d1)

        d3 = self.side3(hx3d)
        d3 = _upsample_like(d3, d1)

        d4 = self.side4(hx4d)
        d4 = _upsample_like(d4, d1)

        d5 = self.side5(hx5d)
        d5 = _upsample_like(d5, d1)

        d6 = self.side6(hx6)
        d6 = _upsample_like(d6, d1)

        d0 = self.outconv(paddle.concat((d1, d2, d3, d4, d5, d6), 1))

        return [d0, d1, d2, d3, d4, d5, d6]

    def init_weight(self):
        if self.pretrained is not None:
            utils.load_entire_model(self, self.pretrained)


class REBNCONV(nn.Layer):
    def __init__(self, in_ch=3, out_ch=3, dirate=1):
        super(REBNCONV, self).__init__()

        self.conv_s1 = nn.Conv2D(
            in_ch, out_ch, 3, padding=1 * dirate, dilation=1 * dirate)
        self.bn_s1 = nn.BatchNorm2D(out_ch)
        self.relu_s1 = nn.ReLU()

    def forward(self, x):

        hx = x
        xout = self.relu_s1(self.bn_s1(self.conv_s1(hx)))

        return xout


## upsample tensor 'src' to have the same spatial size with tensor 'tar'
def _upsample_like(src, tar):

    src = F.upsample(src, size=paddle.shape(tar)[2:], mode='bilinear')

    return src


### RSU-7 ###
class RSU7(nn.Layer):  #UNet07DRES(nn.Layer):
    def __init__(self, in_ch=3, mid_ch=12, out_ch=3):
        super(RSU7, self).__init__()

        self.rebnconvin = REBNCONV(in_ch, out_ch, dirate=1)

        self.rebnconv1 = REBNCONV(out_ch, mid_ch, dirate=1)
        self.pool1 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.rebnconv2 = REBNCONV(mid_ch, mid_ch, dirate=1)
        self.pool2 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.rebnconv3 = REBNCONV(mid_ch, mid_ch, dirate=1)
        self.pool3 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.rebnconv4 = REBNCONV(mid_ch, mid_ch, dirate=1)
        self.pool4 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.rebnconv5 = REBNCONV(mid_ch, mid_ch, dirate=1)
        self.pool5 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.rebnconv6 = REBNCONV(mid_ch, mid_ch, dirate=1)

        self.rebnconv7 = REBNCONV(mid_ch, mid_ch, dirate=2)

        self.rebnconv6d = REBNCONV(mid_ch * 2, mid_ch, dirate=1)
        self.rebnconv5d = REBNCONV(mid_ch * 2, mid_ch, dirate=1)
        self.rebnconv4d = REBNCONV(mid_ch * 2, mid_ch, dirate=1)
        self.rebnconv3d = REBNCONV(mid_ch * 2, mid_ch, dirate=1)
        self.rebnconv2d = REBNCONV(mid_ch * 2, mid_ch, dirate=1)
        self.rebnconv1d = REBNCONV(mid_ch * 2, out_ch, dirate=1)

    def forward(self, x):

        hx = x
        hxin = self.rebnconvin(hx)

        hx1 = self.rebnconv1(hxin)
        hx = self.pool1(hx1)

        hx2 = self.rebnconv2(hx)
        hx = self.pool2(hx2)

        hx3 = self.rebnconv3(hx)
        hx = self.pool3(hx3)

        hx4 = self.rebnconv4(hx)
        hx = self.pool4(hx4)

        hx5 = self.rebnconv5(hx)
        hx = self.pool5(hx5)

        hx6 = self.rebnconv6(hx)

        hx7 = self.rebnconv7(hx6)

        hx6d = self.rebnconv6d(paddle.concat((hx7, hx6), 1))
        hx6dup = _upsample_like(hx6d, hx5)

        hx5d = self.rebnconv5d(paddle.concat((hx6dup, hx5), 1))
        hx5dup = _upsample_like(hx5d, hx4)

        hx4d = self.rebnconv4d(paddle.concat((hx5dup, hx4), 1))
        hx4dup = _upsample_like(hx4d, hx3)

        hx3d = self.rebnconv3d(paddle.concat((hx4dup, hx3), 1))
        hx3dup = _upsample_like(hx3d, hx2)

        hx2d = self.rebnconv2d(paddle.concat((hx3dup, hx2), 1))
        hx2dup = _upsample_like(hx2d, hx1)

        hx1d = self.rebnconv1d(paddle.concat((hx2dup, hx1), 1))

        return hx1d + hxin


### RSU-6 ###
class RSU6(nn.Layer):  #UNet06DRES(nn.Layer):
    def __init__(self, in_ch=3, mid_ch=12, out_ch=3):
        super(RSU6, self).__init__()

        self.rebnconvin = REBNCONV(in_ch, out_ch, dirate=1)

        self.rebnconv1 = REBNCONV(out_ch, mid_ch, dirate=1)
        self.pool1 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.rebnconv2 = REBNCONV(mid_ch, mid_ch, dirate=1)
        self.pool2 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.rebnconv3 = REBNCONV(mid_ch, mid_ch, dirate=1)
        self.pool3 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.rebnconv4 = REBNCONV(mid_ch, mid_ch, dirate=1)
        self.pool4 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.rebnconv5 = REBNCONV(mid_ch, mid_ch, dirate=1)

        self.rebnconv6 = REBNCONV(mid_ch, mid_ch, dirate=2)

        self.rebnconv5d = REBNCONV(mid_ch * 2, mid_ch, dirate=1)
        self.rebnconv4d = REBNCONV(mid_ch * 2, mid_ch, dirate=1)
        self.rebnconv3d = REBNCONV(mid_ch * 2, mid_ch, dirate=1)
        self.rebnconv2d = REBNCONV(mid_ch * 2, mid_ch, dirate=1)
        self.rebnconv1d = REBNCONV(mid_ch * 2, out_ch, dirate=1)

    def forward(self, x):

        hx = x

        hxin = self.rebnconvin(hx)

        hx1 = self.rebnconv1(hxin)
        hx = self.pool1(hx1)

        hx2 = self.rebnconv2(hx)
        hx = self.pool2(hx2)

        hx3 = self.rebnconv3(hx)
        hx = self.pool3(hx3)

        hx4 = self.rebnconv4(hx)
        hx = self.pool4(hx4)

        hx5 = self.rebnconv5(hx)

        hx6 = self.rebnconv6(hx5)

        hx5d = self.rebnconv5d(paddle.concat((hx6, hx5), 1))
        hx5dup = _upsample_like(hx5d, hx4)

        hx4d = self.rebnconv4d(paddle.concat((hx5dup, hx4), 1))
        hx4dup = _upsample_like(hx4d, hx3)

        hx3d = self.rebnconv3d(paddle.concat((hx4dup, hx3), 1))
        hx3dup = _upsample_like(hx3d, hx2)

        hx2d = self.rebnconv2d(paddle.concat((hx3dup, hx2), 1))
        hx2dup = _upsample_like(hx2d, hx1)

        hx1d = self.rebnconv1d(paddle.concat((hx2dup, hx1), 1))

        return hx1d + hxin


### RSU-5 ###
class RSU5(nn.Layer):  #UNet05DRES(nn.Layer):
    def __init__(self, in_ch=3, mid_ch=12, out_ch=3):
        super(RSU5, self).__init__()

        self.rebnconvin = REBNCONV(in_ch, out_ch, dirate=1)

        self.rebnconv1 = REBNCONV(out_ch, mid_ch, dirate=1)
        self.pool1 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.rebnconv2 = REBNCONV(mid_ch, mid_ch, dirate=1)
        self.pool2 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.rebnconv3 = REBNCONV(mid_ch, mid_ch, dirate=1)
        self.pool3 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.rebnconv4 = REBNCONV(mid_ch, mid_ch, dirate=1)

        self.rebnconv5 = REBNCONV(mid_ch, mid_ch, dirate=2)

        self.rebnconv4d = REBNCONV(mid_ch * 2, mid_ch, dirate=1)
        self.rebnconv3d = REBNCONV(mid_ch * 2, mid_ch, dirate=1)
        self.rebnconv2d = REBNCONV(mid_ch * 2, mid_ch, dirate=1)
        self.rebnconv1d = REBNCONV(mid_ch * 2, out_ch, dirate=1)

    def forward(self, x):

        hx = x

        hxin = self.rebnconvin(hx)

        hx1 = self.rebnconv1(hxin)
        hx = self.pool1(hx1)

        hx2 = self.rebnconv2(hx)
        hx = self.pool2(hx2)

        hx3 = self.rebnconv3(hx)
        hx = self.pool3(hx3)

        hx4 = self.rebnconv4(hx)

        hx5 = self.rebnconv5(hx4)

        hx4d = self.rebnconv4d(paddle.concat((hx5, hx4), 1))
        hx4dup = _upsample_like(hx4d, hx3)

        hx3d = self.rebnconv3d(paddle.concat((hx4dup, hx3), 1))
        hx3dup = _upsample_like(hx3d, hx2)

        hx2d = self.rebnconv2d(paddle.concat((hx3dup, hx2), 1))
        hx2dup = _upsample_like(hx2d, hx1)

        hx1d = self.rebnconv1d(paddle.concat((hx2dup, hx1), 1))

        return hx1d + hxin


### RSU-4 ###
class RSU4(nn.Layer):  #UNet04DRES(nn.Layer):
    def __init__(self, in_ch=3, mid_ch=12, out_ch=3):
        super(RSU4, self).__init__()

        self.rebnconvin = REBNCONV(in_ch, out_ch, dirate=1)

        self.rebnconv1 = REBNCONV(out_ch, mid_ch, dirate=1)
        self.pool1 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.rebnconv2 = REBNCONV(mid_ch, mid_ch, dirate=1)
        self.pool2 = nn.MaxPool2D(2, stride=2, ceil_mode=True)

        self.rebnconv3 = REBNCONV(mid_ch, mid_ch, dirate=1)

        self.rebnconv4 = REBNCONV(mid_ch, mid_ch, dirate=2)

        self.rebnconv3d = REBNCONV(mid_ch * 2, mid_ch, dirate=1)
        self.rebnconv2d = REBNCONV(mid_ch * 2, mid_ch, dirate=1)
        self.rebnconv1d = REBNCONV(mid_ch * 2, out_ch, dirate=1)

    def forward(self, x):

        hx = x

        hxin = self.rebnconvin(hx)

        hx1 = self.rebnconv1(hxin)
        hx = self.pool1(hx1)

        hx2 = self.rebnconv2(hx)
        hx = self.pool2(hx2)

        hx3 = self.rebnconv3(hx)

        hx4 = self.rebnconv4(hx3)

        hx3d = self.rebnconv3d(paddle.concat((hx4, hx3), 1))
        hx3dup = _upsample_like(hx3d, hx2)

        hx2d = self.rebnconv2d(paddle.concat((hx3dup, hx2), 1))
        hx2dup = _upsample_like(hx2d, hx1)

        hx1d = self.rebnconv1d(paddle.concat((hx2dup, hx1), 1))

        return hx1d + hxin


### RSU-4F ###
class RSU4F(nn.Layer):  #UNet04FRES(nn.Layer):
    def __init__(self, in_ch=3, mid_ch=12, out_ch=3):
        super(RSU4F, self).__init__()

        self.rebnconvin = REBNCONV(in_ch, out_ch, dirate=1)

        self.rebnconv1 = REBNCONV(out_ch, mid_ch, dirate=1)
        self.rebnconv2 = REBNCONV(mid_ch, mid_ch, dirate=2)
        self.rebnconv3 = REBNCONV(mid_ch, mid_ch, dirate=4)

        self.rebnconv4 = REBNCONV(mid_ch, mid_ch, dirate=8)

        self.rebnconv3d = REBNCONV(mid_ch * 2, mid_ch, dirate=4)
        self.rebnconv2d = REBNCONV(mid_ch * 2, mid_ch, dirate=2)
        self.rebnconv1d = REBNCONV(mid_ch * 2, out_ch, dirate=1)

    def forward(self, x):

        hx = x

        hxin = self.rebnconvin(hx)

        hx1 = self.rebnconv1(hxin)
        hx2 = self.rebnconv2(hx1)
        hx3 = self.rebnconv3(hx2)

        hx4 = self.rebnconv4(hx3)

        hx3d = self.rebnconv3d(paddle.concat((hx4, hx3), 1))
        hx2d = self.rebnconv2d(paddle.concat((hx3d, hx2), 1))
        hx1d = self.rebnconv1d(paddle.concat((hx2d, hx1), 1))

        return hx1d + hxin