PhotosensorToSemanticSegmentation v2
A node that maps object IDs to colors of semantic groups from a SemanticSegmentationLabelConfig
configuration.
Description
This node records commands on the command buffer provided by the custom pass context in InTranscode
that converts a raw lidar frame of type
StructuredBuffer<PhotosensorRawData>
to a ARGB32
texture containing semantic colors.
Semantic segmentation assigns the same color to all instances of the same semantic group.
Inputs
Name | Type | Description |
---|---|---|
FrameHeight | uint |
The vertical frame size. |
Frame | StructuredBuffer<PhotosensorRawData> |
A buffer containing one full sweep from the photosensor. |
InTranscode | CustomPassContext |
The graphic context into which to queue transcoding commands. |
Outputs
Name | Type | Description |
---|---|---|
Segmentation | RenderTexture(Format: ARGB32) |
The segmentation image for the current frame. |
OutTranscode | CustomPassContext |
The graphic context after queuing the transcoding commands. |
Bindings
Name | Type | Description |
---|---|---|
Semantic Label Configuration | SemanticSegmentationLabelConfig |
The mapping between object instance IDs and semantic group colors. |
Execution
Asynchronous. Executes when InTranscode changes.
Notes
This node expects that the size of Frame is a multiple of FrameHeight. If this is not the case, the frame is ignored.
Note
You must install the Perception package in your project to perform lidar segmentation.
Note
You can import the Perception library from the SensorSDK samples to access a lidar prefab and a demo scene demonstrating lidar segmentation.
Warning
Installing the perception package and running it the first time may prompt an error regarding a misconfigured output endpoint. This error has no effect on the lidar segmentation. To remove the error: select Edit > Project Settings > Perception, and change the active endpoint to No Output Endpoint or Perception Endpoint.
Warning
Lidar segmentation requires a PathtracedSensingComponent
. A Camera-Based Lidar solution isn't supported yet.