jamesqfreeman / miceye Goto Github PK

非常棒的工作！感谢开源如此有用的数据采集工具。
但是我有一个疑问，您在论文里写到您使用的是tobii Eye Tracker 4C进行眼动数据采集，那您是否试过使用Tobii Eye Tracker 5来进行眼动数据采集呢？我想了解下使用Tobii Eye Tracker 5进行数据采集的可行性。

因为我在前期调研的时候，发现官方在Tobii5中没有提供有效的数据采集接口，仅是提供了unity环境下的数据采集SDK。所以我在看见您工作之前，一直认为Tobii5可能不能很好的支撑数据采集工作。

期待您的回复！谢谢！

Hi @JamesQFreeman ,

I am working on a similar problem but with a slightly different application. The goal of my project is to identify damage on a structure for visual building inspection. I have collected the data using Tobii Pro Glasses, and 3 of the 10 participants were on two building sites. My data is in the form of videos (3~5 mins long) with all the gaze information included with it.

You have used a screen-based eye tracker for your project, and I have used wearable glasses for my project, so I was wondering if there is a way I can use the data I have collected and replicate it with your method.
Also, while I was going to set up the environment, I got a challenge with Tobii's stream engine API as it's not available publicly, and they only provide a Pro SDK kit. Can you help me this how I can overcome this or how I can simply use my data to use your method and generate attention maps for building inspector's visual attention?

Thank you so much for your help.

When I run python MicEye.py after following the installation tutorial, I get ModuleNotFoundError: No module named 'TobiiEyeTracker'
I tried pip install TobiiEyeTracker, but it didn't work, where can I download this module?

you remind in paper that there are two radiologists in your test. I wonder if the two radiologists look the same X-ray images or different?
if they look the same images, how do you use the one X-ray image with two annotations to caculate the attention consistency loss with uncertainty?