PointEModel<T>
Models & Types · AiDotNet.Diffusion.ThreeD
Point-E model for text-to-3D point cloud generation.
Inheritance: DiffusionModelBase<T> → LatentDiffusionModelBase<T> → ThreeDDiffusionModelBase<T> → PointEModel<T>
For Beginners
Point-E creates 3D objects as “point clouds” - collections of colored points in 3D space that form a shape:
What is a point cloud?
- Thousands of 3D points (X, Y, Z coordinates)
- Each point can have a color (R, G, B)
- Together they form the surface of an object
- Like a very detailed dot-to-dot drawing in 3D
Example: “A red chair”
- Point-E first imagines what the chair looks like (synthetic image)
- Then generates 4096 points forming the chair shape
- Points are colored red where appropriate
- Result: A 3D point cloud you can view from any angle
Use cases:
- 3D modeling: Quick prototypes for games, VR, AR
- Visualization: Create 3D representations from descriptions
- Dataset creation: Generate synthetic 3D training data
How It Works
Point-E is OpenAI’s model for generating 3D point clouds from text descriptions. It uses a two-stage pipeline: first generating a synthetic view of the object, then generating a point cloud conditioned on that view.
Technical specifications:
- Default point count: 4096 (can generate 1024, 4096, or 16384)
- Coordinate range: [-1, 1] normalized
- Color: RGB values [0, 1]
- Two-stage: Image generation + point cloud diffusion
- Inference: ~40 steps for image, ~64 for point cloud
Example
// Create a Point-E model
var pointE = new PointEModel<float>();
// Generate a point cloud from text
var pointCloud = pointE.GeneratePointCloud(
prompt: "A wooden chair",
numPoints: 4096,
numInferenceSteps: 64,
guidanceScale: 3.0);
// pointCloud shape: [1, 4096, 6] - XYZ + RGB per point
// Generate from an image
var image = ImageHelper<float>.LoadImage("chair_photo.jpg");
var fromImage = pointE.GenerateFromImage(image, numPoints: 4096);
// Export to PLY file for viewing in 3D software
// pointCloud is the result tensor — write it out with your own encoder.
Constructors
| Constructor | Summary |
|---|---|
PointEModel(NeuralNetworkArchitecture<T> architecture, DiffusionModelOptions<T> options, INoiseScheduler<T> scheduler, DiTNoisePredictor<T> pointCloudPredictor, ILatentDiffusionModel<T> imageGenerator, IConditioningModule<T> conditioner, int defaultPointCount, bool useTwoStage, int? seed) |
Initializes a new Point-E model with full customization support. |
Properties
| Property | Summary |
|---|---|
DiTNoisePredictor<T> PointCloudPredictor { get; } |
Gets the point cloud predictor. |
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; } |
|
bool UsesTwoStage { get; } |
Gets whether this model uses two-stage generation. |
int LatentChannels { get; } |
Methods
| Method | Summary |
|---|---|
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> GenerateFromImage(Tensor<T> image, int? numPoints, int numInferenceSteps, int? seed) |
Generates a point cloud from an image. |
Tensor<T> GeneratePointCloud(string prompt, string negativePrompt, int? numPoints, int numInferenceSteps, double guidanceScale, int? seed) |
Generates a colored point cloud from a text prompt. |
ValueTuple<Tensor<T>, Tensor<T>> ConvertToMesh(Tensor<T> pointCloud, int resolution) |
Converts point cloud to mesh using marching cubes (simplified). |