Typed data and results catalog¶
Typed values make composition safe. They prevent an array on a reduced lattice from being mistaken for a canonical ERP and keep numeric values, geometric support and data validity distinct.
Registered typed-data surfaces¶
Surface |
Stability |
Canonical import |
Contract |
Public components or entries |
|---|---|---|---|---|
spherical-region-analysis-data |
Stable |
panorai.data |
panorai-spherical-region-analysis/v1 |
EnclosingSphericalCap, RegionAdjacencyEdge, RegionAdjacencyGraph, RegionAnalysisResult, RegionMeasurement, RegionMeasurements, SphericalBoundary, SphericalBoundarySet, SphericalContour, SphericalContourSet, panorai.data.regions.EnclosingSphericalCap, panorai.data.regions.RegionAdjacencyEdge, panorai.data.regions.RegionAdjacencyGraph, panorai.data.regions.RegionAnalysisResult, panorai.data.regions.RegionMeasurement, panorai.data.regions.RegionMeasurements, panorai.data.regions.SphericalBoundary, panorai.data.regions.SphericalBoundarySet, panorai.data.regions.SphericalContour, panorai.data.regions.SphericalContourSet |
spherical-observations |
Stable |
panorai.data |
panorai-spherical-observations/v1 |
AggregatedViewEvidence, DataFactory, EquirectangularImage, FacePointBatch, FaceRegionBatch, FaceViewOutput, GnomonicFace, GnomonicFaceSet, ObservationFusionResult, RasterFieldResult, RasterFieldSpec, SphericalBinaryMask, SphericalPointObservation, SphericalPointSet, SphericalRegionObservation, SphericalRegionSet, SphericalTranspose, SphericalViewObservation, SphericalViewSet, XYZImageDataset, panorai.data.utils.PanoraiError, panorai.data.utils.InvalidDataError, panorai.data.utils.ChannelMismatchError, panorai.data.utils.MetadataValidationError, panorai.data.utils.DataConversionError, panorai.data.utils.MissingChannelError, panorai.data.utils.ensure_type_consistency, panorai.data.utils.ShapeManager, panorai.data.xyz_image_dataset.XYZImageDataset, panorai.data.xyz_image_dataset.extract_hw_from_ply_filename, panorai.data.xyz_image_dataset.read_encrypted_ply_xyz_image, panorai.data.xyz_image_dataset.read_ply_xyz_image |
spherical-core-contracts |
Experimental |
panorai.core |
panorai-core-operations/v1 |
SPHERICAL_CORE_INTERFACE, SphericalConv2dSpec, SphericalExecutionSpec, SphericalGridSpec, SphericalNeighborhoodSpec, SphericalPool2dSpec, SphericalResizeSpec, panorai.core.specs.BackendName, panorai.core.specs.Interpolation, panorai.core.specs.Layout, panorai.core.specs.SphericalConv2dSpec, panorai.core.specs.SphericalExecutionSpec, panorai.core.specs.SphericalGridSpec, panorai.core.specs.SphericalNeighborhoodSpec, panorai.core.specs.SphericalPool2dSpec, panorai.core.specs.SphericalResizeSpec |
panorama-stitching-data |
Experimental |
panorai.stitching |
panorai-panorama-stitching/v1-experimental |
ImagePairMatches, ImagePairRegistration, PanoramaStitchingResult, RegisteredImageSet, StitchingDiagnostics, StitchingFeatureSet, StitchingFeatures, StitchingImage, StitchingInput, StitchingMatches, StitchingOptions, WarpedImageSet |
Object workflow¶
Component |
Represents |
Key responsibility |
|---|---|---|
|
one ERP with one or more fields |
Image-style immutable I/O and transforms backed by canonical spherical geometry. |
|
one projected view |
Values, projection spec, support, validity and source identity. |
|
ordered projected views |
Per-view mapping, typed fields and reconstruction. |
|
named spherical orientation transform |
Explicit alternative to ambiguous planar transpose conventions. |
Pipeline values¶
Component |
Represents |
Carries |
|---|---|---|
|
spatial identity of a raster |
domain, shape, ERP reference, anchor, stride, periodicity, view identity and transform chain. |
|
one named raster modality |
values, kind, units, support, validity and |
|
named heterogeneous branches |
raster and non-raster results without forcing one common topology. |
|
callable-output declaration |
kind, units, invalid policy and minimum valid weight. |
|
one Stage transition |
output value plus JSON-friendly execution details. |
|
recorded Stage execution |
name, identity, details and provenance used by |
|
complete recorded execution |
final output, retained values, ordered records and manifest metadata. |
Per-view and spherical observations¶
Component |
Topology |
Use |
|---|---|---|
|
one view |
Retain face ID, projection, support and validity with a non-raster result. |
|
ordered views |
Generic keypoints, detections or another declared per-view kind. |
|
score vector per view |
Aggregate or retain classifications without inventing a raster grid. |
|
embedding per view |
Preserve feature vectors, source fields and view support. |
|
lifted observations on the sphere |
Bearings, regions or evidence after an explicit lifting step. |
|
points, regions or view evidence |
Geometry-aware reductions such as angular NMS, solid-angle overlap and coverage aggregation. |
Feature and pose values¶
Component |
Produced by |
Meaning |
|---|---|---|
|
application input adapter |
Ordered reference/source ERP pair and optional IDs. |
|
|
Ordered feature sets before matching. |
|
|
Matched features, validity and provenance. |
|
|
Weighted unit bearings in both panorama frames. |
|
spherical relative-pose estimator |
Source-from-reference |
|
estimator diagnostics |
Residual, coverage, parallax, cheirality, stability and model-competition evidence. |
Dense stereo returns SphericalStereoResult: radial range, confidence,
validity and diagnostics. Semantic applications return the segmentation and
proposal types listed in the complete public inventory.
Dense reconstruction values¶
RegisteredSphericalView and RegisteredSphericalScene bind central ERP
images to supplied metric world poses. MultiViewRangeMap keeps range,
sampling support, observability, validity and source evidence distinct;
MultiViewRangeSet groups those maps without treating confidence as a
cross-reference probability. TSDFVolume stores ZYX values, weights and
view counts with an XYZ world origin. SurfaceMesh stores world XYZ
vertices, faces, positive-TSDF normals, per-vertex evidence and optional
explicitly valid color. SphericalDenseReconstructionResult records either
accepted or rejected and retains intermediates according to policy.
Region-analysis results¶
The Stable panorai.data region types are
RegionMeasurements/RegionMeasurement, RegionAdjacencyGraph and
RegionAdjacencyEdge, SphericalBoundarySet/SphericalBoundary,
SphericalContourSet/SphericalContour, EnclosingSphericalCap and
their combined RegionAnalysisResult. Arrays are copied read-only.
Boundaries are explicitly unordered pixel-centre bearing samples; contours are
ordered cell-edge bearings. Neither is mislabeled as a filled polygon.
Registration and stitching values¶
The Experimental panorai.stitching data flow is explicit from input to
result:
Component |
Produced by |
Carries |
|---|---|---|
|
Application input adapter |
Planar values, independent support and validity, image identity and optional horizontal FOV. |
|
Feature and overlap Stages |
Per-image pixels/descriptors and matched candidate image pairs. |
|
Calibration and registration Stage |
Pairwise evidence, intrinsics, camera-to-world SO(3) rotations, rejected pairs and optimization costs. |
|
Warp and photometric Stages |
Per-source ERP values, support, validity, weights and gains. |
|
Blend Stage or |
Partial typed ERP, contribution/support maps and complete diagnostics. |
All arrays are copied read-only. A result does not imply full-sphere coverage; the panorama masks and contribution evidence remain authoritative.
See spherical observations and EquirectangularImage for detailed contracts.