// Available camera render outputs for Real-Time Path-Tracing (RT2) mode. Use when user asks what AOVs/render vars are available, what format or dtype an output has, or how to read a specific output like depth, normals, albedo, or distance.
High-level overview of a typical ovrtx application lifecycle. Use when user asks how to structure an ovrtx program, what the main steps are, or how the pieces fit together.
Asynchronous operation patterns including polling, timeouts, and non-blocking workflows. Use when user asks about async rendering, non-blocking operations, polling, timeouts, or parallel rendering.
Creating persistent attribute bindings for efficient repeated writes to the same prims and attribute. Use when user asks about persistent bindings, repeated writes, efficient animation loops, bind_attribute, or updating transforms every frame with caller-owned tensors. Use mapping-attributes when the hot path needs zero-copy direct writes into ovrtx buffers.
Binding materials to prims at runtime. Use when user asks to assign a material, change a material, set material binding, or swap materials on a prim.
Cloning USD subtrees to create copies at new paths. Use when user asks to clone, duplicate, copy a prim, or create instances of existing geometry.
Authoring and configuring OmniLidar sensor prims and lidar PointCloud render outputs. Use when user asks to create a lidar scene, configure an OmniLidar prim, choose lidar output frame/coordinate behavior, or request lidar PointCloud channels.
| name | camera-outputs-rt2 |
| description | Available camera render outputs for Real-Time Path-Tracing (RT2) mode. Use when user asks what AOVs/render vars are available, what format or dtype an output has, or how to read a specific output like depth, normals, albedo, or distance. |
| license | LicenseRef-NvidiaProprietary |
| version | 0.3.0 |
| author | NVIDIA ovrtx |
| tags | ["ovrtx","rendering","camera"] |
| tools | ["Read","Grep"] |
Use this skill when the user asks what AOVs/render vars are available, what format or dtype an output has, or how to read a specific output like depth, normals, albedo, or distance.
Resolve inputs in this order: existing repository files and referenced snippets, explicit user request, then broader agent context.
> **Source:** snippet before writing or explaining API usage.reading-render-output for the concrete map/read code once the requested camera output is selected.LdrColor unless the request specifically needs HDR, depth, normals, semantic IDs, or another listed AOV.RenderVar entries to the RenderProduct orderedVars; each extra output has memory cost.sourceName, dtype, shape, and readback skill needed for the requested output.This skill has no scripts.
DepthSD currently returns all zeros through the C API.Most applications only need LdrColor unless the user specifically requests other outputs (e.g. depth, normals). Do not add extra AOVs by default -- each one has a memory cost.
The following RenderVar sourceName values are available when using Real-Time Path-Tracing mode. Set each as the sourceName attribute on a RenderVar prim under a RenderProduct.
sourceName | Format | Type | NumPy dtype | Shape | Description |
|---|---|---|---|---|---|
LdrColor | RGBA | uint8 | uint8 | (H, W, 4) | Tone-mapped sRGB color. Standard display-ready output. |
HdrColor | RGBA | float16 | float16 | (H, W, 4) | Linear-space HDR color. Use for post-processing and compositing. |
NormalSD | XYZA | float32 | float32 | (H, W, 4) | World-space surface normals. |
DepthSD | Z | float32 | float32 | (H, W, 1) | Unitless depth, mapped from 1 (near clip) to 0 (far clip). Known bug: returns all zeros via the C API. |
DistanceToCameraSD | Z | float32 | float32 | (H, W, 1) | Euclidean distance from camera origin to surface, in meters. |
DistanceToImagePlaneSD | Z | float32 | float32 | (H, W, 1) | Perpendicular distance from image plane to surface, in meters. |
DiffuseAlbedoSD | RGBA | uint8 | uint8 | (H, W, 4) | Diffuse surface albedo (base color without lighting). |
Camera3dPositionSD | XYZA | float32 | float32 | (H, W, 4) | Camera-space 3D position of each surface point, in scene units. |
SemanticSegmentation | ID | uint32 | uint32 | (H, W, 1) | Per-pixel semantic ID. Decode SemanticIdMap to map IDs to semantic label strings. |
SemanticIdMap | IdentifierMap | uint8 buffer | uint8 | buffer | Packed metadata buffer mapping semantic IDs to label strings. |
Source:
tests/docs/python/test_camera_aovs.pysnippetdoc-camera-aov-smoke-test
In C, image DLTensor outputs use channel-last shape (H, W, C) with dtype.lanes = 1. The dtype.code reflects the format kind directly: kDLUInt for LdrColor / DiffuseAlbedoSD / SemanticSegmentation, kDLFloat for HdrColor / NormalSD / DepthSD / DistanceToCameraSD / DistanceToImagePlaneSD / Camera3dPositionSD. SemanticIdMap is a packed uint8 metadata buffer, not an image.
Source:
tests/docs/c/test_camera_aovs.cppsnippetdoc-camera-aov-smoke-test-c
Add each output as a RenderVar prim under a RenderProduct, referencing it
in orderedVars:
Source:
tests/docs/python/test_camera_aovs.pysnippetdoc-camera-aov-usda
Only include the outputs you need -- there is no requirement to request all of them.
reading-render-output -- how to map and access render var pixel dataconfiguring-lidar-sensors / configuring-radar-sensors for non-camera sensors, and docs/sensors/configuration.rst for RenderProduct and RenderVar prim setupstepping-and-rendering -- stepping the renderer and iterating over resultsRenderVar is listed in the parent RenderProduct orderedVars.> **Source:** directives in this skill to locate tested snippets before reusing API patterns.