SnapVideoModel<T>
Models & Types · AiDotNet.Diffusion.Video.LongVideo
Snap Video scaled spatiotemporal transformer for text-to-video.
Inheritance: DiffusionModelBase<T> → LatentDiffusionModelBase<T> → VideoDiffusionModelBase<T> → SnapVideoModel<T>
For Beginners
SnapVideo from Snap uses a scaled spatiotemporal transformer that processes space and time jointly for high-quality video generation. Its efficient architecture scales well to longer durations.
How It Works
References:
- Paper: “Snap Video: Scaled Spatiotemporal Transformers for Text-to-Video Synthesis” (Snap, 2024)
Snap Video uses a scaled spatiotemporal transformer architecture for efficient text-to-video synthesis. The model scales transformer attention across both spatial and temporal dimensions with efficient factorization strategies. Achieves strong quality with efficient generation.
Technical specifications:
- Architecture: Scaled Spatiotemporal Transformer
- Latent channels: 16
- Default: 48 frames at 24 FPS
- Supports I2V: Yes | T2V: Yes | V2V: No
Example
var options = new DiffusionModelOptions<float> { LatentChannels = 16, DefaultInferenceSteps = 50 };
var noise = Tensor<float>.CreateRandom(new[] { 1, 16, 48, 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 SnapVideoModel<float>(options: options))
.Build(trainX, trainY);
var video = result.Predict(noise);
Constructors
| Constructor | Summary |
|---|---|
SnapVideoModel(NeuralNetworkArchitecture<T> architecture, DiffusionModelOptions<T> options, INoiseScheduler<T> scheduler, DiTNoisePredictor<T> predictor, TemporalVAE<T> temporalVAE, IConditioningModule<T> conditioner, int defaultNumFrames, int defaultFPS, int? seed) |
Initializes a new instance of SnapVideoModel with full customization support. |
Properties
| Property | Summary |
|---|---|
IConditioningModule<T> Conditioner { get; } |
|
INoisePredictor<T> NoisePredictor { get; } |
|
IVAEModel<T> TemporalVAE { get; } |
|
IVAEModel<T> VAE { get; } |
|
bool SupportsImageToVideo { get; } |
|
bool SupportsTextToVideo { get; } |
|
bool SupportsVideoToVideo { get; } |
|
int LatentChannels { get; } |
Methods
| Method | Summary |
|---|---|
ModelMetadata<T> GetModelMetadata() |