opencd/models/backbones/tinycd.py

"""
Codegoni A, Lombardi G, Ferrari A. 
TINYCD: A (Not So) Deep Learning Model For Change Detection[J]. 
arXiv preprint arXiv:2207.13159, 2022.
The code in this file is borrowed from:
https://github.com/AndreaCodegoni/Tiny_model_4_CD
"""
from typing import List, Optional

import torchvision
from torch import Tensor, reshape, stack
from torch.nn import (Conv2d, InstanceNorm2d, Module, ModuleList, PReLU,
                      Sequential, Upsample)

from opencd.registry import MODELS


class PixelwiseLinear(Module):
    def __init__(
        self,
        fin: List[int],
        fout: List[int],
        last_activation: Module = None,
    ) -> None:
        assert len(fout) == len(fin)
        super().__init__()

        n = len(fin)
        self._linears = Sequential(
            *[
                Sequential(
                    Conv2d(fin[i], fout[i], kernel_size=1, bias=True),
                    PReLU()
                    if i < n - 1 or last_activation is None
                    else last_activation,
                )
                for i in range(n)
            ]
        )

    def forward(self, x: Tensor) -> Tensor:
        # Processing the tensor:
        return self._linears(x)


class MixingBlock(Module):
    def __init__(
        self,
        ch_in: int,
        ch_out: int,
    ):
        super().__init__()
        self._convmix = Sequential(
            Conv2d(ch_in, ch_out, 3, groups=ch_out, padding=1),
            PReLU(),
            InstanceNorm2d(ch_out),
        )

    def forward(self, x: Tensor, y: Tensor) -> Tensor:
        # Packing the tensors and interleaving the channels:
        mixed = stack((x, y), dim=2)
        mixed = reshape(mixed, (x.shape[0], -1, x.shape[2], x.shape[3]))

        # Mixing:
        return self._convmix(mixed)


class MixingMaskAttentionBlock(Module):
    """use the grouped convolution to make a sort of attention"""

    def __init__(
        self,
        ch_in: int,
        ch_out: int,
        fin: List[int],
        fout: List[int],
        generate_masked: bool = False,
    ):
        super().__init__()
        self._mixing = MixingBlock(ch_in, ch_out)
        self._linear = PixelwiseLinear(fin, fout)
        self._final_normalization = InstanceNorm2d(ch_out) if generate_masked else None
        self._mixing_out = MixingBlock(ch_in, ch_out) if generate_masked else None

    def forward(self, x: Tensor, y: Tensor) -> Tensor:
        z_mix = self._mixing(x, y)
        z = self._linear(z_mix)
        z_mix_out = 0 if self._mixing_out is None else self._mixing_out(x, y)

        return (
            z
            if self._final_normalization is None
            else self._final_normalization(z_mix_out * z)
        )


class UpMask(Module):
    def __init__(
        self,
        scale_factor: float,
        nin: int,
        nout: int,
    ):
        super().__init__()
        self._upsample = Upsample(
            scale_factor=scale_factor, mode="bilinear", align_corners=True
        )
        self._convolution = Sequential(
            Conv2d(nin, nin, 3, 1, groups=nin, padding=1),
            PReLU(),
            InstanceNorm2d(nin),
            Conv2d(nin, nout, kernel_size=1, stride=1),
            PReLU(),
            InstanceNorm2d(nout),
        )

    def forward(self, x: Tensor, y: Optional[Tensor] = None) -> Tensor:
        x = self._upsample(x)
        if y is not None:
            x = x * y
        return self._convolution(x)


def _get_backbone(
    bkbn_name, pretrained, output_layer_bkbn, freeze_backbone
) -> ModuleList:
    # The whole model:
    entire_model = getattr(torchvision.models, bkbn_name)(
        pretrained=pretrained
    ).features

    # Slicing it:
    derived_model = ModuleList([])
    for name, layer in entire_model.named_children():
        derived_model.append(layer)
        if name == output_layer_bkbn:
            break

    # Freezing the backbone weights:
    if freeze_backbone:
        for param in derived_model.parameters():
            param.requires_grad = False
    return derived_model


@MODELS.register_module()
class TinyCD(Module):
    def __init__(
        self,
        in_channels,
        bkbn_name="efficientnet_b4",
        pretrained=True,
        output_layer_bkbn="3",
        freeze_backbone=False,
    ):
        super().__init__()

        # Load the pretrained backbone according to parameters:
        self._backbone = _get_backbone(
            bkbn_name, pretrained, output_layer_bkbn, freeze_backbone
        )

        # Initialize mixing blocks:
        self._first_mix = MixingMaskAttentionBlock(6, 3, [3, 10, 5], [10, 5, 1])
        self._mixing_mask = ModuleList(
            [
                MixingMaskAttentionBlock(48, 24, [24, 12, 6], [12, 6, 1]),
                MixingMaskAttentionBlock(64, 32, [32, 16, 8], [16, 8, 1]),
                MixingBlock(112, 56),
            ]
        )

        # Initialize Upsampling blocks:
        self._up = ModuleList(
            [
                UpMask(2, 56, 64),
                UpMask(2, 64, 64),
                UpMask(2, 64, 32),
            ]
        )

        # Final classification layer:
        self._classify = PixelwiseLinear([32, 16], [16, 1], None) # out_channels = 8

    def forward(self, x1: Tensor, x2: Tensor) -> Tensor:
        features = self._encode(x1, x2)
        latents = self._decode(features)
        out = self._classify(latents)
        
        return (out,)

    def _encode(self, ref, test) -> List[Tensor]:
        features = [self._first_mix(ref, test)]
        for num, layer in enumerate(self._backbone):
            ref, test = layer(ref), layer(test)
            if num != 0:
                features.append(self._mixing_mask[num - 1](ref, test))
        return features

    def _decode(self, features) -> Tensor:
        upping = features[-1]
        for i, j in enumerate(range(-2, -5, -1)):
            upping = self._up[i](upping, features[j])
        return upping