Instant3DModel<T>

Models & Types · AiDotNet.Diffusion.ThreeD

Instant3D model – fast text-to-3D with feed-forward generation.

Inheritance: DiffusionModelBase<T> → LatentDiffusionModelBase<T> → ThreeDDiffusionModelBase<T> → Instant3DModel<T>

For Beginners

Instant3D creates 3D from text in under 1 second:

Key characteristics:

  • Feed-forward: no per-shape optimization needed
  • Multi-view generation + instant reconstruction
  • Sub-second 3D generation
  • NeRF output with mesh extraction

Reference: Li et al., “Instant3D: Fast Text-to-3D with Sparse-View Generation and Large Reconstruction Model”, ICLR 2024

How It Works

Instant3D generates 3D objects from text in a single forward pass using a multi-view diffusion model with a feed-forward reconstruction network.

Example

var instant3d = new Instant3DModel<float>();
var mesh = instant3d.GenerateMesh(
    prompt: "A wooden treasure chest",
    resolution: 256,
    numInferenceSteps: 50,
    guidanceScale: 7.5,
    seed: 42);

Constructors

Constructor Summary
Instant3DModel(NeuralNetworkArchitecture<T> architecture, DiffusionModelOptions<T> options, INoiseScheduler<T> scheduler, UNetNoisePredictor<T> unet, StandardVAE<T> vae, IConditioningModule<T> conditioner, int defaultPointCount) Initializes a new instance of Instant3DModel with full customization support.

Properties

Property Summary
IConditioningModule<T> Conditioner { get; }
INoisePredictor<T> NoisePredictor { get; }
IVAEModel<T> VAE { get; }
bool SupportsMesh { get; }
bool SupportsNovelView { get; }
bool SupportsPointCloud { get; }
bool SupportsScoreDistillation { get; }
bool SupportsTexture { get; }
int LatentChannels { get; }

Methods

Method Summary
Mesh3D<T> GenerateMesh(string prompt, string negativePrompt, int resolution, int numInferenceSteps, double guidanceScale, int? seed)
Mesh3D<T> ImageTo3D(Tensor<T> inputImage, int numViews, int numInferenceSteps, double guidanceScale, int? seed) Not supported: image-to-3D is not part of this model’s published pipeline.
ModelMetadata<T> GetModelMetadata()
Tensor<T> GeneratePointCloud(string prompt, string negativePrompt, int? numPoints, int numInferenceSteps, double guidanceScale, int? seed)