Tool Tracking Postprocessor Operator

Authors: Holoscan Team (NVIDIA)
Supported platforms: x86_64, aarch64
Language: C++, Python
Last modified: March 9, 2026
Latest version: 1.0
Minimum Holoscan SDK version: 0.5.0
Tested Holoscan SDK versions: 0.5.0
Contribution metric: Level 1 - Highly Reliable

The tool_tracking_postprocessor extension provides a codelet that converts inference output of lstm_tensor_rt_inference used in the endoscopy tool tracking pipeline to be consumed by the holoviz codelet.

nvidia::holoscan::tool_tracking_postprocessor

Tool tracking postprocessor codelet

Parameters

• in: Input channel, type gxf::Tensor
• type: gxf::Handle<gxf::Receiver>
• out: Output channel, type gxf::Tensor
• type: gxf::Handle<gxf::Transmitter>
• min_prob: Minimum probability, (default: 0.5)
• type: float
• overlay_img_colors: Color of the image overlays, a list of RGB values with components between 0 and 1, (default: 12 qualitative classes color scheme from colorbrewer2)
• type: std::vector<std::vector<float>>
• device_allocator: Output Allocator
• type: gxf::Handle<gxf::Allocator>
• cuda_stream_pool: Instance of gxf::CudaStreamPool
• type: gxf::Handle<gxf::CudaStreamPool>

API Reference

Python

ToolTrackingPostprocessorOp

Constructor Parameters

Parameter	Type	Required	Description
fragment	Fragment	Required	The fragment that the operator belongs to.
device_allocator	holoscan.resources.Allocator	Required	Output allocator used on the device side.
min_prob	float	Optional	Minimum probability (in range [0, 1]). Default value is 0.5.
overlay_img_colors	sequence of sequence of float	Optional	Color of the image overlays, a list of RGB values with components between 0 and 1. The default value is a qualitative colormap with a sequence of 12 colors.
cuda_stream_pool	holoscan.resources.CudaStreamPool	Optional	holoscan.resources.CudaStreamPool instance to allocate CUDA streams. Default value is None.
name	str	Optional	The name of the operator.