A Python package for streaming Gaze data from EyeTechDS devices.
- Windows OS
- XP / 7 / 8 / 10 (32 or 64 bit)
- The EyeTechDS SDK
- A valid driver for your device. On Windows 10, the "EyeTech HID Eye Tracking Device" driver is automatically installed upon connecting the device.
Supported devices (non-exhaustive)
Install from source
$ pip install git+https://github.com/intheon/eyetechds-lsl
To begin the LSL stream (Gaze) from the first available EyeTechDS device:
$ eyetechlsl stream
To specify a configuration file:
$ eyetechlsl stream --config device-config.file
If you wish to also start a VideoRaw stream:
$ eyetechlsl stream --video
Ensure that you have opencv-python installed.
$ eyetechlsl video
The VideoRaw stream will transmit at the same rate as the Gaze stream, but a new VideoRaw sample will only be transmitted when the QuickLink2 API receives a new frame, so actual frames-per-second will likely be lower. On a VT-3 Mini running at full resolution, actual FPS was around 6-7.
Device configuration uses the format and parameters specified in ./config.template.
By default a ./config file will be loaded if it exists.
EyeTechDS devices prioritize streaming processed data before raw video data. As a result, the VideoRaw stream DeviceDownsampleScaleFactor configuration parameter (float between 1.0 - 20.0) specifies the fractional downscaling of the video data that the device will transmit. A higher DeviceDownsampleScaleFactor value will result in an increase in frames-per-second, at the cost of reduced image resolution.
Lab Streaming Layer or LSL is a system designed to unify the collection of time series data for research experiments. It has become standard in the field of EEG-based brain-computer interfaces for its ability to make seperate streams of data available on a network with time synchronization and near real-time access. For more information, check out this lecture from Modern Brain-Computer Interface Design or the LSL repository
This open-source LSL application was developed and is maintained by Intheon (www.intheon.io).
For support inquiries please file a GitHub issue on this repo (preferred) or contact [email protected].
Copyright (C) 2020-2021 Syntrogi Inc. dba Intheon.
This work was funded by the Army Research Laboratory under Cooperative Agreement Number W911NF-10-2-0022.