UNet3D<T>

Models & Types · AiDotNet.NeuralNetworks

Represents a 3D U-Net neural network for volumetric semantic segmentation.

Inheritance: NeuralNetworkBase<T> → DeclaredModelLayoutBase<T> → VolumetricModelLayoutBase<T> → UNet3D<T>

For Beginners

A 3D U-Net is like an intelligent 3D scanner that can identify and label every single voxel (3D pixel) in a 3D volume.

Think of it like this:

  • The encoder (left side of “U”) looks at the big picture by progressively zooming out
  • The decoder (right side of “U”) zooms back in to produce detailed predictions
  • Skip connections (horizontal lines in “U”) preserve fine details from encoder to decoder

This is useful for:

  • Medical imaging: Finding organs or tumors in CT/MRI scans
  • 3D scene understanding: Segmenting objects in point clouds
  • Part segmentation: Identifying different parts of 3D shapes

The “U” shape comes from the symmetric encoder-decoder design with skip connections.

How It Works

A 3D U-Net extends the classic U-Net architecture to three dimensions for processing volumetric data. It uses an encoder-decoder structure with skip connections to produce dense, per-voxel predictions while preserving both local details and global context.

Example

var architecture = new NeuralNetworkArchitecture<float>(inputFeatures: 8, outputSize: 4);
var volume = Tensor<float>.CreateRandom(new[] { 1, 1, 64, 64, 64 });
var trainX = Tensor<float>.CreateRandom(4, 8);
var trainY = Tensor<float>.CreateRandom(4, 2);
var result = new AiModelBuilder<float, Tensor<float>, Tensor<float>>()
    .ConfigureModel(new UNet3D<float>(architecture))
    .Build(trainX, trainY);
var segmented = result.Predict(volume);

Constructors

Constructor Summary
UNet3D() Initializes a new instance with default architecture settings.
UNet3D(NeuralNetworkArchitecture<T> architecture, IGradientBasedOptimizer<T, Tensor<T>, Tensor<T>> optimizer, ILossFunction<T> lossFunction, UNet3DOptions options) Initializes the model from an already-resolved options instance.

Properties

Property Summary
int BaseFilters { get; set; } Gets the base number of filters in the first encoder block.
int NumClasses { get; set; } Gets the number of output classes (segmentation categories).
int NumEncoderBlocks { get; set; } Gets the number of encoder blocks in the network.
int VoxelResolution { get; set; } Gets the voxel grid resolution used by this network.

Methods

Method Summary
ModelMetadata<T> GetModelMetadata() Gets metadata about this model for serialization and inspection.
Tensor<T> Forward(Tensor<T> input) Performs a forward pass through the network.
Tensor<T> ForwardForTraining(Tensor<T> input) Training must take the same skip path as inference.