Skip to content

Supported Hardware

jaredjstein edited this page Jan 16, 2018 · 21 revisions

Supported Imagers

RGB-D Imagers currently supported:

Legacy:

N.B.: Legacy imagers should still function with OpenPtrack v2, however driver updates are not supported by the OpenPTrack team.

Supported Hardware

The main requirement for each compute node: Linux-compatible x86 hardware capable of running Ubuntu 16.04 and the Robot Operating System (ROS). Most likely, one compute node will be needed per imager. Our deployment experience suggests that modern Ubuntu-compatible machines with i7 CPUs and at least 8 GB RAM constitute a good setup for each node. Running on pre-2011 hardware is not recommended. OpenPTrack will not compile on a machine with less than 2.7GB RAM.

Faster CPU(s) can mean better tracking performance, and may also enable running more than one detection process (i.e., more than one sensor) per node. Do not run any node close to 100% CPU utilization, or tracking performance suffer will most likely suffer.

Kinect v2 support requires a CUDA-capable GPU. The NVIDIA GPUs that have been tested so far: NVidia GeForce 650, 660, 670, 740, 750, 760, 770, 840, 850, 860, 870, 1070, and 1080.

N.B.: Pose annotation requires a NVidia 1070 or 1080 chip set with the 1080 being recommend.

The Deployment Guide contains additional hardware recommendations, in the Tested Hardware section.

Supported OS

  • Person and object detection and tracking & pose annotation hosts: Ubuntu 16.04.
  • Consumers: Any platform that can receive UDP datagrams.

Setting Up an OpenPTrack v2 System:

Running OpenPTrack v2:

Tracking GUI

How to receive tracking data in:

  1. Tested Hardware
  2. Network Configuration
  3. Imager Mounting and Placement
  4. Calibration in Practice
  5. Quick Start Example
  6. Imager Settings
  7. Manual Ground Plane
  8. Calibration Refinement (Person-Based)
  9. Calibration Refinement (Manual)

OPT on the NVidia Jetson

Clone this wiki locally