models/unet.py

"""
UNet model definition for lane segmentation.
Includes DoubleConv block and UNet architecture.
"""
import torch
import torch.nn as nn
import torch.nn.functional as F

# --- DoubleConv Block ---
class DoubleConv(nn.Module):
    """
    (Conv => BN => ReLU) * 2 block used in UNet encoder/decoder.
    Args:
        in_channels: Number of input channels
        out_channels: Number of output channels
    """
    def __init__(self, in_channels: int, out_channels: int):
        """
        Args:
            in_channels (int): Number of input channels
            out_channels (int): Number of output channels
        """
        super().__init__()
        self.double_conv = nn.Sequential(
            nn.Conv2d(in_channels, out_channels, kernel_size=3, padding=1),
            nn.BatchNorm2d(out_channels),
            nn.ReLU(inplace=True),
            nn.Conv2d(out_channels, out_channels, kernel_size=3, padding=1),
            nn.BatchNorm2d(out_channels),
            nn.ReLU(inplace=True)
        )

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        """
        Forward pass for DoubleConv block.
        Args:
            x (torch.Tensor): Input tensor
        Returns:
            torch.Tensor: Output tensor
        """
        return self.double_conv(x)

# --- UNet Model ---
class UNet(nn.Module):
    """
    U-Net: Convolutional Networks for Biomedical Image Segmentation
    Args:
        in_channels: Number of input channels
        out_channels: Number of output channels
    """
    def __init__(self, in_channels: int = 3, out_channels: int = 1):
        """
        Args:
            in_channels (int): Number of input channels
            out_channels (int): Number of output channels
        """
        super().__init__()
        self.encoder1 = DoubleConv(in_channels, 64)
        self.pool1 = nn.MaxPool2d(kernel_size=2)
        self.encoder2 = DoubleConv(64, 128)
        self.pool2 = nn.MaxPool2d(kernel_size=2)
        self.encoder3 = DoubleConv(128, 256)
        self.pool3 = nn.MaxPool2d(kernel_size=2)
        self.encoder4 = DoubleConv(256, 512)
        self.pool4 = nn.MaxPool2d(kernel_size=2)
        self.bottleneck = DoubleConv(512, 1024)
        self.upconv4 = nn.ConvTranspose2d(1024, 512, kernel_size=2, stride=2)
        self.decoder4 = DoubleConv(1024, 512)
        self.upconv3 = nn.ConvTranspose2d(512, 256, kernel_size=2, stride=2)
        self.decoder3 = DoubleConv(512, 256)
        self.upconv2 = nn.ConvTranspose2d(256, 128, kernel_size=2, stride=2)
        self.decoder2 = DoubleConv(256, 128)
        self.upconv1 = nn.ConvTranspose2d(128, 64, kernel_size=2, stride=2)
        self.decoder1 = DoubleConv(128, 64)
        self.final_conv = nn.Conv2d(64, out_channels, kernel_size=1)

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        """
        Forward pass for UNet model.
        Args:
            x (torch.Tensor): Input tensor
        Returns:
            torch.Tensor: Output tensor
        """
        """
        Forward pass of UNet.
        Args:
            x: Input tensor of shape (B, C, H, W)
        Returns:
            Output tensor of shape (B, out_channels, H, W)
        """
        enc1 = self.encoder1(x)
        enc2 = self.encoder2(self.pool1(enc1))
        enc3 = self.encoder3(self.pool2(enc2))
        enc4 = self.encoder4(self.pool3(enc3))
        bottleneck = self.bottleneck(self.pool4(enc4))
        dec4 = self.upconv4(bottleneck)
        dec4 = torch.cat([dec4, enc4], dim=1)
        dec4 = self.decoder4(dec4)
        dec3 = self.upconv3(dec4)
        dec3 = torch.cat([dec3, enc3], dim=1)
        dec3 = self.decoder3(dec3)
        dec2 = self.upconv2(dec3)
        dec2 = torch.cat([dec2, enc2], dim=1)
        dec2 = self.decoder2(dec2)
        dec1 = self.upconv1(dec2)
        dec1 = torch.cat([dec1, enc1], dim=1)
        dec1 = self.decoder1(dec1)
        out = self.final_conv(dec1)
        return torch.sigmoid(out)

# --- Model Summary and FLOPs (Optional) ---
if __name__ == "__main__":
    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
    model = UNet(in_channels=3, out_channels=1).to(device)
    dummy_input = torch.randn(1, 3, 256, 256).to(device)
    output = model(dummy_input)
    print(f"Output shape: {output.shape}")

    # Model summary
    from torchinfo import summary
    print(summary(model, input_size=(1, 3, 256, 256), device=device))

    # FLOPs and parameters
    from thop import profile
    flops, params = profile(model, inputs=(dummy_input,))
    print(f"FLOPs: {flops:,}")
    print(f"Parameters: {params:,}")