KO | EN

GitLite

Minimalist Open Source Repository Search Engine

gitlite — search

#dsh #dsh-plugin #ai-agent #developer-tools #codex #typescript #cli #deepseek-harness #claude-code #ai-agents #mcp #rust

isaac_ros_benchmark

★ 35 Open GitHub ↗

Performance benchmarking for NVIDIA-accelerated Isaac ROS packages

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Isaac ROS Benchmark

Performance benchmarking for NVIDIA-accelerated Isaac ROS packages.

## Overview

Isaac ROS Benchmark builds upon the
[ros2_benchmark](https://github.com/NVIDIA-ISAAC-ROS/ros2_benchmark)
to provide configurations to benchmark Isaac ROS graphs. Performance
results that measure Isaac ROS for throughput, latency, and utilization
enable robotics developers to make informed decisions when designing
real-time robotics applications. The Isaac ROS performance results can
be independently verified, as the method, configuration, and data input
used for benchmarking are provided.

The `ros2_benchmark` playback node plug-in, for type adaptation and
negotiation, is provided for
[NITROS](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_nitros), which
optimizes the performance of message transport costs through
[RCL](https://github.com/ros2/rclcpp) with GPU accelerated graphs of
nodes.

The datasets for benchmarking are explicitly not downloaded by default.
To pull down the standardized benchmark datasets, refer to the
[ros2_benchmark Dataset](https://github.com/NVIDIA-ISAAC-ROS/ros2_benchmark/blob/release-4.5/README.md#datasets)
section.

---

## Documentation

Please visit the [Isaac ROS Documentation](https://nvidia-isaac-ros.github.io/repositories_and_packages/isaac_ros_benchmark/index.html) to learn how to use this repository.

---

## Packages

* [`isaac_ros_benchmark`](https://nvidia-isaac-ros.github.io/repositories_and_packages/isaac_ros_benchmark/isaac_ros_benchmark/index.html)
  * [Quickstart](https://nvidia-isaac-ros.github.io/repositories_and_packages/isaac_ros_benchmark/isaac_ros_benchmark/index.html#quickstart)
  * [Troubleshooting](https://nvidia-isaac-ros.github.io/repositories_and_packages/isaac_ros_benchmark/isaac_ros_benchmark/index.html#troubleshooting)
* [`isaac_ros_moveit_benchmark`](https://nvidia-isaac-ros.github.io/repositories_and_packages/isaac_ros_benchmark/benchmarks/isaac_ros_moveit_benchmark/index.html)
  * [Overview](https://nvidia-isaac-ros.github.io/repositories_and_packages/isaac_ros_benchmark/benchmarks/isaac_ros_moveit_benchmark/index.html#overview)
  * [Setup](https://nvidia-isaac-ros.github.io/repositories_and_packages/isaac_ros_benchmark/benchmarks/isaac_ros_moveit_benchmark/index.html#setup)
  * [Running the Benchmark](https://nvidia-isaac-ros.github.io/repositories_and_packages/isaac_ros_benchmark/benchmarks/isaac_ros_moveit_benchmark/index.html#running-the-benchmark)
  * [Available Metrics](https://nvidia-isaac-ros.github.io/repositories_and_packages/isaac_ros_benchmark/benchmarks/isaac_ros_moveit_benchmark/index.html#available-metrics)

## Latest

Update 2026-04-30: Compatibility and integration updates for the Isaac ROS 4.4.0 release

## Performance

| Node | Input Size | AGX Thor T5000 | AGX Thor T4000 | DGX Spark | x86_64 w/ RTX 5090 |
|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| [AprilTag Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_apriltag_benchmark/scripts/isaac_ros_apriltag_node.py) | 720p | [385 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_apriltag_node-agx_thor.json) 2.9 ms @ 30Hz | [280 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_apriltag_node-thor-t4000.json) 11 ms @ 30Hz | [462 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_apriltag_node-dgx_spark.json) 2.4 ms @ 30Hz | [596 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_apriltag_node-x86-5090.json) 2.0 ms @ 30Hz |
| [FoundationPose Pose Estimation Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_foundationpose_benchmark/scripts/isaac_ros_foundationpose_node.py) | 720p | [3.15 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_foundationpose_node-agx_thor.json) 370 ms @ 30Hz | [3.40 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_foundationpose_node-thor-t4000.json) 350 ms @ 30Hz | [2.77 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_foundationpose_node-dgx_spark.json) 360 ms @ 30Hz | [10.1 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_foundationpose_node-x86-5090.json) 99 ms @ 30Hz |
| [Stereo Disparity Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_stereo_image_proc_benchmark/scripts/isaac_ros_disparity_node.py) | 1080p | [217 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_disparity_node-agx_thor.json) 5.7 ms @ 30Hz | [196 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_disparity_node-thor-t4000.json) 20 ms @ 30Hz | [249 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_disparity_node-dgx_spark.json) 5.4 ms @ 30Hz | [747 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_disparity_node-x86-5090.json) 4.0 ms @ 30Hz |
| [Rectify Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_image_proc_benchmark/scripts/isaac_ros_rectify_node.py) | 1080p | [1550 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_rectify_node-agx_thor.json) 1.6 ms @ 30Hz | [768 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_rectify_node-thor-t4000.json) 4.8 ms @ 30Hz | [2500 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_rectify_node-dgx_spark.json) 2.3 ms @ 30Hz | [2500 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_rectify_node-x86-5090.json) 2.9 ms @ 30Hz |
| [TensorRT Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_tensor_rt_benchmark/scripts/isaac_ros_tensor_rt_dope_node.py) DOPE | VGA | [165 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_tensor_rt_dope_node-agx_thor.json) 6.4 ms @ 30Hz | [147 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_tensor_rt_dope_node-thor-t4000.json) 27 ms @ 30Hz | [107 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_tensor_rt_dope_node-dgx_spark.json) 8.6 ms @ 30Hz | [300 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_tensor_rt_dope_node-x86-5090.json) 4.0 ms @ 30Hz |
| [Triton Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_triton_benchmark/scripts/isaac_ros_triton_dope_node.py) DOPE | VGA | [161 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_triton_dope_node-agx_thor.json) 7.5 ms @ 30Hz | [153 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_triton_dope_node-thor-t4000.json) 30 ms @ 30Hz | [94.5 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_triton_dope_node-dgx_spark.json) 10 ms @ 30Hz | [281 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_triton_dope_node-x86-5090.json) 4.4 ms @ 30Hz |
| [TensorRT Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_tensor_rt_benchmark/scripts/isaac_ros_tensor_rt_ps_node.py) PeopleSemSegNet | 544p | [566 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_tensor_rt_ps_node-agx_thor.json) 2.1 ms @ 30Hz | [407 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_tensor_rt_ps_node-thor-t4000.json) 6.8 ms @ 30Hz | [594 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_tensor_rt_ps_node-dgx_spark.json) 1.4 ms @ 30Hz | [1570 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_tensor_rt_ps_node-x86-5090.json) 1.5 ms @ 30Hz |
| [Triton Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_triton_benchmark/scripts/isaac_ros_triton_ps_node.py) PeopleSemSegNet | 544p | [361 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_triton_ps_node-agx_thor.json) 4.0 ms @ 30Hz | [280 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_triton_ps_node-thor-t4000.json) 11 ms @ 30Hz | [362 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_triton_ps_node-dgx_spark.json) 3.1 ms @ 30Hz | [485 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_triton_ps_node-x86-5090.json) 5.4 ms @ 30Hz |
| [DNN Image Encoder Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_dnn_image_encoder_benchmark/scripts/isaac_ros_dnn_image_encoder_node.py) | VGA | [356 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_dnn_image_encoder_node-agx_thor.json) 11 ms @ 30Hz | [300 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_dnn_image_encoder_node-thor-t4000.json) 19 ms @ 30Hz | [318 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_dnn_image_encoder_node-dgx_spark.json) 10 ms @ 30Hz | [417 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_dnn_image_encoder_node-x86-5090.json) 10 ms @ 30Hz |
| [Occupancy Grid Localizer Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_occupancy_grid_localizer_benchmark/scripts/isaac_ros_grid_localizer_node.py) | ~50 sq. m | [28.1 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_grid_localizer_node-agx_thor.json) 40 ms @ 30Hz | [21.1 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_grid_localizer_node-thor-t4000.json) 110 ms @ 30Hz | [40.1 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_grid_localizer_node-dgx_spark.json) 27 ms @ 30Hz | [50.1 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_grid_localizer_node-x86-5090.json) 13 ms @ 30Hz |
| [H.264 Decoder Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_h264_decoder_benchmark/scripts/isaac_ros_h264_decoder_node.py) | 1080p | [296 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_h264_decoder_node-agx_thor.json) 4.5 ms @ 30Hz | [296 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_h264_decoder_node-thor-t4000.json) 5.3 ms @ 30Hz | [736 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_h264_decoder_node-dgx_spark.json) 4.4 ms @ 30Hz | [296 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_h264_decoder_node-x86-5090.json) 4.1 ms @ 30Hz |
| [H.264 Encoder Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_h264_encoder_benchmark/scripts/isaac_ros_h264_encoder_iframe_node.py) I-frame Support | 1080p | [296 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_h264_encoder_iframe_node-agx_thor.json) 2.9 ms @ 30Hz | [297 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_h264_encoder_iframe_node-thor-t4000.json) 10 ms @ 30Hz | [921 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_h264_encoder_iframe_node-dgx_spark.json) 2.7 ms @ 30Hz | [1050 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_h264_encoder_iframe_node-x86-5090.json) 3.8 ms @ 30Hz |
| [H.264 Encoder Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_h264_encoder_benchmark/scripts/isaac_ros_h264_encoder_pframe_node.py) P-frame Support | 1080p | [296 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_h264_encoder_pframe_node-agx_thor.json) 2.9 ms @ 30Hz | [297 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_h264_encoder_pframe_node-thor-t4000.json) 9.5 ms @ 30Hz | [846 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_h264_encoder_pframe_node-dgx_spark.json) 3.1 ms @ 30Hz | [1050 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_h264_encoder_pframe_node-x86-5090.json) 3.8 ms @ 30Hz |

| Graph | Input Size | AGX Thor T5000 | AGX Thor T4000 | DGX Spark | x86_64 w/ RTX 5090 |
|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| [AprilTag Graph](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_apriltag_benchmark/scripts/isaac_ros_apriltag_graph.py) | 720p | [354 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_apriltag_graph-agx_thor.json) 5.2 ms @ 30Hz | [273 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_apriltag_graph-thor-t4000.json) 15 ms @ 30Hz | [546 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_apriltag_graph-dgx_spark.json) 3.9 ms @ 30Hz | [596 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_apriltag_graph-x86-5090.json) 2.8 ms @ 30Hz |
| [Centerpose Pose Estimation Graph](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_centerpose_benchmark/scripts/isaac_ros_centerpose_graph.py) | VGA | [50.2 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_centerpose_graph-agx_thor.json) 27 ms @ 30Hz | [50.4 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_centerpose_graph-thor-t4000.json) 73 ms @ 30Hz | [50.2 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_centerpose_graph-dgx_spark.json) 23 ms @ 30Hz | [50.2 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_centerpose_graph-x86-5090.json) 19 ms @ 30Hz |
| [DOPE Pose Estimation Graph](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_dope_benchmark/scripts/isaac_ros_dope_graph.py) | VGA | [143 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_dope_graph-agx_thor.json) 22 ms @ 30Hz | [116 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_dope_graph-thor-t4000.json) 53 ms @ 30Hz | [89.3 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_dope_graph-dgx_spark.json) 27 ms @ 30Hz | [193 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_dope_graph-x86-5090.json) 17 ms @ 30Hz |
| [DNN Stereo Disparity Graph](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_ess_benchmark/scripts/isaac_ros_ess_graph.py) Full | 576p | [157 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_ess_graph-agx_thor.json) 13 ms @ 30Hz | [68.5 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_ess_graph-thor-t4000.json) 38 ms @ 30Hz | [150 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_ess_graph-dgx_spark.json) 13 ms @ 30Hz | [347 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_ess_graph-x86-5090.json) 9.9 ms @ 30Hz |
| [DNN Stereo Disparity Graph](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_ess_benchmark/scripts/isaac_ros_light_ess_graph.py) Light | 288p | [285 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_light_ess_graph-agx_thor.json) 9.2 ms @ 30Hz | [155 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_light_ess_graph-thor-t4000.json) 20 ms @ 30Hz | [231 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_light_ess_graph-dgx_spark.json) 11 ms @ 30Hz | [349 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_light_ess_graph-x86-5090.json) 9.2 ms @ 30Hz |
| [Stereo Disparity Graph](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_stereo_image_proc_benchmark/scripts/isaac_ros_disparity_graph.py) | 1080p | [203 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_disparity_graph-agx_thor.json) 10 ms @ 30Hz | [185 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_disparity_graph-thor-t4000.json) 26 ms @ 30Hz | [183 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_disparity_graph-dgx_spark.json) 9.0 ms @ 30Hz | [600 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_disparity_graph-x86-5090.json) 6.9 ms @ 30Hz |
| [DetectNet Object Detection Graph](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_detectnet_benchmark/scripts/isaac_ros_detectnet_graph.py) | 544p | [143 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_detectnet_graph-agx_thor.json) 18 ms @ 30Hz | [111 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_detectnet_graph-thor-t4000.json) 43 ms @ 30Hz | [88.0 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_detectnet_graph-dgx_spark.json) 19 ms @ 30Hz | [227 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_detectnet_graph-x86-5090.json) 18 ms @ 30Hz |
| [Grounding DINO Object Detection Graph](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_grounding_dino_benchmark/scripts/isaac_ros_grounding_dino_graph.py) | 544p | [23.4 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_grounding_dino_graph-agx_thor.json) 50 ms @ 30Hz | [22.3 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_grounding_dino_graph-thor-t4000.json) 90 ms @ 30Hz | [17.5 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_grounding_dino_graph-dgx_spark.json) 63 ms @ 30Hz | [130 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_grounding_dino_graph-x86-5090.json) 15 ms @ 30Hz |
| [RT-DETR Object Detection Graph](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_rtdetr_benchmark/scripts/isaac_ros_rtdetr_graph.py) SyntheticaDETR | 720p | [188 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_rtdetr_graph-agx_thor.json) 12 ms @ 30Hz | [156 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_rtdetr_graph-thor-t4000.json) 35 ms @ 30Hz | [165 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_rtdetr_graph-dgx_spark.json) 12 ms @ 30Hz | [444 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_rtdetr_graph-x86-5090.json) 11 ms @ 30Hz |
| [TensorRT Graph](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_unet_benchmark/scripts/isaac_ros_unet_graph.py) PeopleSemSegNet | 544p | [449 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_unet_graph-agx_thor.json) 8.1 ms @ 30Hz | [319 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_unet_graph-thor-t4000.json) 19 ms @ 30Hz | [562 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_unet_graph-dgx_spark.json) 6.7 ms @ 30Hz | [1330 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_unet_graph-x86-5090.json) 6.3 ms @ 30Hz |
| [SAM Image Segmentation Graph](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_segment_anything_benchmark/scripts/isaac_ros_segment_anything_graph.py) Full SAM | 720p | [2.26 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_sam_graph-agx_thor.json) 350 ms @ 30Hz | [2.24 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_sam_graph-thor-t4000.json) 290 ms @ 30Hz | [2.22 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_sam_graph-dgx_spark.json) 280 ms @ 30Hz | [20.8 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_sam_graph-x86-5090.json) 57 ms @ 30Hz |
| [SAM Image Segmentation Graph](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/benchmarks/isaac_ros_segment_anything_benchmark/scripts/isaac_ros_mobile_segment_anything_graph.py) Mobile SAM | 720p | [15.0 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_mobile_sam_graph-agx_thor.json) 230 ms @ 30Hz | [15.0 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_mobile_sam_graph-thor-t4000.json) 200 ms @ 30Hz | [14.6 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_mobile_sam_graph-dgx_spark.json) 82 ms @ 30Hz | [70.3 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/release-4.5/results/isaac_ros_mobile_sam_graph-x86-5090.json) 20 ms @ 30Hz |

isaac_ros_benchmark

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README