Neural Radiance Fields

20 public types in this namespace, organized by kind.

Models & Types (12)

Type Summary
AdaptiveDensificationSchedule Adaptive schedule (excellence goal): fires when the caller-observed loss plateaus below a threshold OR the gradient-norm variance settles — signals that the cloud has stabilized enough that split noise won’t destabilize training, while its…
ExifIntrinsics EXIF intrinsics bundle used to derive per-photo focal-in-pixels when the loader was populated from a camera JPEG.
FixedIntervalDensificationSchedule Fixed-interval schedule — reference-impl behavior.
GaussianSplatting<T> Implements 3D Gaussian Splatting for real-time novel view synthesis.
ImageView<T> A single photo + camera pose used as input to image-space (photometric) radiance-field training (#1834).
InstantNGP<T> Implements Instant Neural Graphics Primitives (Instant-NGP) for fast NeRF training and rendering.
IsotropicPrior<T> Paper-standard fallback prior — synthesizes the anchor view’s mean color at every pixel (spatially uniform).
NeRF<T> Implements Neural Radiance Fields (NeRF) for novel view synthesis.
PixelBatch<T> A batch of pixels sampled from one or more ImageView views — the target side of image-space radiance-field training (#1834).
ProgressiveSamplingSchedule Progressive coarse-to-fine ray-sampling schedule (#1834 excellence goal #4).
Ray<T> Represents a ray in 3D space for rendering.
SceneBounds Auto-derived scene bounds + near/far — output of SceneBoundsEstimator.

Base Classes (1)

Type Summary
LearnedPrior<T> Optional single-image / few-shot reconstruction prior attached to an ImageView.

Interfaces (2)

Type Summary
IDensificationSchedule Densification firing schedule for GaussianSplatting (#1835 excellence goal #2).
IRadianceField<T> Defines the core functionality for neural radiance field models.

Options & Configuration (1)

Type Summary
ImageTrainingOptions Options bundle for image-space (photometric) radiance-field training (#1834 excellence goals #3 + #4).

Helpers & Utilities (4)

Type Summary
AiModelResultRadianceFieldExtensions Model-specific inference extensions on AiModelResult for the radiance-field family (NeRF, InstantNGP, GaussianSplatting).
ImageTrainingDataLoaders
RenderingMetrics<T> Provides image quality metrics for evaluating neural rendering methods like NeRF.
SceneBoundsEstimator Estimates a tight scene bounding box + near/far ray sampling bounds from a set of camera poses (#1834 excellence goal #3).