-
Notifications
You must be signed in to change notification settings - Fork 65
System Requirements
When setting up an OpenPTrack system, you will need, at a minimum, an imager and a computer to connect to this imager. However, to take full advantage of OpenPTrack's features, multiple imagers with Nvidia CUDA graphics cards are needed, as is consumer-grade networking equipment. For example, the hardware required for an OpenPTrack system with three Kinect V2s that can run person tracking, object tracking, and pose annotation, and can utilize YOLO:
- (3) - Computers with i7 Intel processors;
- (3) - Nvidia 1070 or 1080 graphic processors (for pose annotation), or other NVidia GPUs for Kinect V2 integration;
- (3) - Microsoft Kinect V2;
- (1) - Router;
- (1) - Networking Switch (if needed).
This is in addition to any interconnects needed such as Cat 5e/6 cables, power cabling, etc.
A person and object tracking system would be similar to the system outlined above, but the GPU requirements are less constrained. Any Nvidia GPU with an architecture of 870 or above will suffice for person and object tracking.
- System Requirements
- Supported Hardware
- Initial Network Configuration
- Example Hardware List for UCLA Setup
- Making the Checkerboard
- Time Synchronization
- Pre-Tracking Configuration
- Camera Network Configuration
- Single Camera
- Setting Parameters
- Multi-Sensor Person Tracking
- HOG vs YOLO Detectors
- World Coordinate Settings
- Single Camera
- Pose Initialization
- Multi Sensor Pose Annotation
- Pose Best Practices
- Setting Parameters
- Single Camera
- Setting Parameters
- Multi Sensor Object Tracking
- YOLO Custom Training & Testing
- Yolo Trainer
- Single Camera
- Setting Parameters
- Multi Sensor Face Detection and Recognition
- Face Detection and Recognition Data Format
How to receive tracking data in: