Hello neon Team. Are you working with some sport teams or individuals wit using the neon pupil lab for screening. Do you have some ideas where can pupil lab help to screen or improve in the differente sport?

Dear Nadia You are right. I have just an idea and think to start using the Neon for these purposes. And I will be happy if you already have some experiences and ideas and can share with me.

@user-51c732 Neon is a great option if you want to collect eye tracking data in sports applications. We have some videos on our website showing Neon when playing basketball/riding a bike etc.

For sports applications, we recommend opting for the frame "Ready Set Go" frame. This frame has an adjustable headband and fits well under helmets and other headwear.  It is suitable for sports and other activities involving fast movements, ensuring the module stays firmly on the subject’s head.

If you could share more details about your planned research, that will help me provide better feedback. We can also schedule a Demo call if you prefer and we can show you Neon in action 😉

Thank you very much It will be great to have a call and go through

Ok

Hello, why does Neon stop recording a couple minutes into my experiment?

Hi everyone, We're interested in ordering a frame (a ready set go) as an extra for our recently purchased neon device. Before we do so, I would like to ask if there is any possibility we could request some customization of the headband on it?

Sorry I just noticed on the website that headbands are also available. Is it possible to get some images of the head straps (beyond those on the web page, i'm specifically hoping to get a closer look at the earpiece sleeve) ? and would it be possible to vary their length if needed? lastly, is there any room for customization of the strap material (elastic band instead of sheathed steel cable)?

Hola!, Lamentablemente sufri el robo del movil que viene con mi equipo Neon. Que dispositivo me recomienda adquirir para utilizar mis lentes

Moto Edge 50 Fusion es una opción?

Hola @user-fd453d , Lamento escuchar que te hayan robado el teléfono. Puedes ver los dispositivos compatibles en este enlace: https://docs.pupil-labs.com/neon/hardware/compatible-devices/. Personalmente, te recomendaría el Motorola Edge 40 en alguna de sus variantes, ya que es el dispositivo más potente.

Moto Edge 50 es una opción?

Solamente podemos garantizar, dado que han sido probados, el correcto funcionamiento con los dispositivos listados anteriormente.

ok gracias

Hi folks, we're using the ready set go frame with our neon device and within 10 minutes of starting to recording, a good bit of heat that can be felt on the skin. While i expect electronics to generate heat, its a lot more than I anticipated and its getting quite hot 15 minutes in. Is it expected to get this hot? If so, does anyone have any suggestions for how to deal with wearer discomfort if we're recording over hours?

Hi @user-d407c1. Thank you. We have the ready set go frame. It has a plastic housing on the top which seems to have a grating kind of construction near a coil on the electronics package (i'm assuming for the purpose of heat dissipation).

I want to know what the amplitude of an eye movement in degrees (i.e., as an angle from the center of the glasses essentially). The simplest way seems to be to get the coordinates of a square in the scene camera exported by the pupil neon (which is the info I have), and work out the ratio with the estimated angular dimensions of the same square. Is this reasonable, or is there a better way?

Pupil Cloud Add-ons Q&A

I apologize if this is a silly question, but is there a way/option to use Neon without wifi (on the phone) and upload the recordings locally instead of through Cloud? Thank you!

What OS version is requried to run for the Neon glasses?

of the phone

oneplus 10 pro

its running 14.0 Right now

Okay thanks

Hey pupil labs, i wrote a code that reads gaze data and world timestamp data from two CSV files, calculates relative time based on the first timestamp from the world timestamp file, and then converts the relative time in nanoseconds to a Minutes : Seconds : Milli Seconds format.

I happen to see that the converted file says that i have some gaze data beyond the time of the recording. Ex: In my case the recording end at 17minutes:31seconds but the last value in converted gaze file is 17minutes:59seconds:339ms.

What might be the problem

Hi everyone,

I’m using the Neon eye-tracking glasses for a study where participants look at their own bodies in a mirror. I need to define AOIs for each participant (e.g., hands, legs, stomach), and I want the software to automatically extract the eye-tracking data from the video of each participant looking at the mirror, based on these AOIs.

I understand that I need to have a reference image, but I’ve also seen mention of a scanning video. Is the scanning video absolutely necessary for this process? I’ve come across studies that don’t seem to use a scanning video, so I’m unsure of its role here.

Could someone provide clear instructions on how to do this correctly? I’ve managed to get it working in a way, but I’m not confident it’s the most systematic approach for research purposes. A detailed, step-by-step guide would be greatly appreciated!

Thanks in advance!

hello, is saccade detection (+ export) only available in pupil cloud or also neon player? can't seem to find it in player 😦

lucky me 🙂 great thx!!

(y) thx

awesome! thx

I have a question about best practices when using tag aligner (https://github.com/pupil-labs/tag-aligner). For the first step in tag-aligner it says you have to make a RIM enrichment. For this step you need a scanning video of the scene, a still photograph of the scene, and the actual eye tracking recording that is of interest. The github readme for the tag aligner mentions that at least one of these two recordings needs to include the AprilTag. Currently I am including the AprilTag in the scanning video (since it cannot be included in the actual eye tracking recording because of the experiment). However, does the still reference image that is used for the RIM step also need to contain the AprilTag? Also will the RIM enrichment that is used by the tag aligner be less accurate because the scene changes between the scanning and eye tracking recording since the AprilTag is not present in both?

Hi Neon team, I noticed there's a new feature in the Neon Companion app: Companion Settings --> NeonNet --> Compute eye state after a recent app update. I didn't remember seeing the "Compute eye state" option from the older versions. So is it related to the realtime eye state feature introduced in this April? So is this option only for generating the 3D eye state in REAL TIME (like the realtime gaze)? I tried with this feature both turned on and off in the app, and found no difference in the timeseries data downloaded from the Pupil Cloud. Even with this feature turned off, the Pupil Cloud will still generate 3d_eye_states.csv. Can you give me a use case where we need to turn on this feature? Thanks!

Hey pupil labs, one of my recording shows a error like this, it says it will get fixed. how long will it aproximately take ?

Hi Pupil Labs! We are using the neon glasses indoor in our lab. We are using it for a car simulator that we built. We have two problems, and one question: - At the start of the recording(usually for 3sec, but in extreme cases 10-15 sec), the whole screen is grey, only the red circle(gaze data) is shown. I will include a screenshot about this. -The second problem is about overheating. After a few recordings, the module in the middle gets really hot, and it affects the performance of the glasses. -Question: Is there any way to connect the glasses directly to a windows machine, and manage the functions from there? (Recording, downloading videos directly) Thank you for your help!

blinks.csv timestamps alignment

Hello! I am working on achieving the precise time sync for my devices. the documentation was very clear and the code worked beautifully (thank you for all the work put into this!). However, I have a question about the output. The first time I ran the code the offset estimate was about 37ms, and we are trying to achieve an offset of <20ms max. Thus, i restarted my computer and the mobile phone devices, and toggled the phones to update their time. After this, the new output was -950ms (even worse!). Any ideas why this might be occurring and what I can do to achieve an offset of less than 20ms?

our aim is to have two neon eyeglasses recording at the exact same time. having the precise time is essential for ensuring that we can have both eye glasses recorded at the exact same time

both would be ideal

@user-87d763 Ok, I see. And just one last question: do you mean that you want both Neons to collect a gaze datum, for example, at exactly or close to exactly the same timestamp?

Hey everyone, We're encountering an issue with the synchronization of the timestamps between the frames and IMU data. Specifically, we've noticed a delay of approximately 1:30 minutes between the two datasets in some sessions. After performing calibration movements during the session, we tried to synchronize the data by looking at the frame numbers marking the start and end of the movements, retrieving the corresponding timestamps, and then extracting the IMU data within that time frame, but found that the frame timestamps didn't align with the corresponding IMU data. After investigating, we found that the start times of the IMU and the frames were off by about 1:30 minutes. When we shifted the IMU data by this amount the data matched again.

However, this issue has only occurred in certain sessions—while the sessions before and after these had no such issues. Additionally, both sensors were shut down with the usual delay of 1-2 seconds at the end of each session, as expected.

In the plot below, the first panel shows the IMU data with the original timestamps, with the red lines marking the frame timestamps. Furthermore, approximately 1:30 minutes after the frame timestamps, we can also see the expected pattern in the IMU data. In the bottom panel, the IMU timestamps are shifted, and you can see the expected pattern in the IMU data between the frame timestamps.

Has anyone encountered something similar or have ideas on what might be causing this? Any help or suggestions would be greatly appreciated! Thanks in advance!

Hi, we use neon player to process data after recording. We record each trial separately so, when we go to use Neon Player, we have to drag the export file into the interface and then wait for processing to finish. With many trials, manually uploading, waiting, then downloading results is time consuming. Is there any way to load multiple recordings for processing at once or to do them in succession automatically?

Hola tengo claro los equipos recomendados por pupil, sufrimos el robo del equipo mobil y tenemos un samsung A55, con android 14 pero no reconoce las gafas. Alguna chance de ocupar este equipo?...en chile no se comercializa la marca one plus y el moto 40 no se encuentra en el mercado. Que me recomiendan?

Hola @user-fd453d - Siento mucho que te hayan robado el móvil. Lamentablemente, los únicos dispositivos compatibles son los recomendados en nuestra página web (OnePlus8T, OnePlus10 y Moto Edge 40 pro). Si no puedes encontrar un teléfono compatible de un distribuidor local, podemos ofrecerte uno; por favor, ponte en contacto a través del correo electrónico [email removed] y el equipo de ventas te proporcionará más detalles.

nadia consulta, y el motorola edge 50 fusion o edge 50 pro?

Hola nadia solo para confirmar, este es el equipo compatible. Es un motorola edge 40, no es edge 40 pro.

Hi Pupil team, regarding to the recent decision of limiting the cloud storage, I was wondering whether if I download the data from the cloud and remove it, would it be possible to upload the data to the cloud back. It would be useful if I can upload the data after I remove it from the cloud in case I want to do some post-processing using the cloud system that I didn't realize beforehand. Thanks for your response in advance!

Team I need some urgent help. My device is not getting recognised on the phone

I have an open ticket. Can someone please join and help me with it

The device is getting recognised as a charging device instead of the glasses.

Hi everybody, I‘m trying to Transfer Recordings in the Pupil Cloud from one Workspace to another workspace from a different person. I want them to be displayed just like their own recordings in their workspace. I don’t just want to invite the person to my workspace. Is there a way to make that possible? Kind Regards Nicola

Hi @user-da6260 , it is not possible to transfer recordings between Workspaces. Are you experiencing errors with the Workspace invite process?

Hello. I put a USB HUB on the smartphone with an HDMI output. I can display Neon on the TV. But I cannot display app in landscape. How must I do this ?

The solid colors of the walls aren't going to be helpful for RIM at all, so anything you can do there to add more detail / (visual) texture would help... but I'd bet that what you have there is enough if the floor is constantly in view. With things like this though, the most honest answer is often "try it and see"

Hi is this an appropriate place to ask a question about AOI's and heat maps?

Thank you @user-cdcab0 and @nmt . I think I might print some visual texture for the walls to be on the safe side. Is there a way of validating the accuracy of a RIM enrichment for the 'try it and see' method? Or is it just a case of genetrated a heat map of an enriched eye tracking recording and checking that it looks 'right'?

I have never updated my Motorola Edge 40 Pro that came with the bundle, although it keeps prompting me to update. I believe the instructions are not to update, as the companion app is designed to run optimally for the phone OS as it came in the package. I was wondering if the new companion app announced requires some updates, or we should keep the OS as it is?

Hey guys. Has anyone noticed any network interference between the Neon streaming via the unity package or through the web neon.local:8080 page? I have an image generator program running for our simulator, and I lose connection to the Neon when its running. I checked with wireshark and they are using completely different ports for communication, and there's a fair bit of traffic (60 pps for the image generator, and whatever I set the streaming to for the neon), but I feel like that isn't the issue. Any ideas?

Hello! I have a very simple question; Is there a way to toggle off the fixation lines? I only want a video with the numbered Fixation points, no lines connecting them. Thanks!

(apologies if this should go in feature requests plz let me know)

Hello @user-f43a29 and others at Pupil Labs. We have a Vuzix Z100 smart glasses we're using for an AR study at the MIT Media Lab. We'd like to integrate the https://pupil-labs.com/products/neon/shop#bare-metal Bare Metal Neon with it - do you think that's possible?

Thank you! I did try the ref image mapper last year, but I look at a lot of paintings in one visit. I found that to take a scan video each time interfered too much with my process. The video is good enough to illustrate for my purposes at the moment, especially since the points are transformed further in my plugin for c4d 🙂 Thank you for the recommendation though!

Hi Team, If we have audio recording enabled is the audio generally not audible during the playback on the phone itself?

Is the microphone on the glasses or does it use the mic from the recording device?

Neon Microphones & Audio playback

Hey pupil labs, One of my recording is showing like this ? how long should i wait for it to get processed ??

Hi! When I download gaze data from Neon Player, the cvs file contains timestamps and gaze coordinates. How do I get the corresponding frame numbers?

hello, I'm having troubles with the pupil neon. I'm connecting them to the companion device, I can see a white led turning on on the glasses but in the companion app the device is not recognised and there is still written plug in and go. What does that means?

Have a question re: lighting. I've done a scene recording for manual mapping (it's too complex and dynamic for RIM), but unfortunately the lighting was very poor due to the nature of the environment, and hence it's difficult to make out specific features for MM to map it accurately

Is there any way to boost the brightness within the mapper video?

I'm also getting an "internal server error" + "fixation not found" when trying to perform manual mapping on a couple of videos, they have approx 12000 and 9000 fixations. Another video with about 2000 fixations isn't giving this error.

One quick question. I am attempting to connect to the IP Address and stream the view…but unfortunately the streaming does not work regardless of my browser or computer that is used. Are there any specifics to try besides?! Or ways to reset the Ip address or URL?

Hi @user-3bcaa4 , which eyetracker do you have? It sounds like you might have Pupil Invisible or Neon?

Neon

Yes they are connected to the same connection. And it’s a work connection. Usually with a mobile hotspot, so that could also be impacting the streaming I’m sure as well. It did work previously though using a hotspot for a brief time though

Good day a quick question is it possible to get the eye tracking videos (as in the pupil eye videos) off the https://cloud.pupil-labs.com/

Hi, we have been using neon to record for a few months now with no issues. The last couple of times that we tried, the glasses were not recognized by the companion device. The light on the front of the glasses turns on while plugged in, but the app doesn't show that it's connected and we can't record. We are using the given USBC cable. Is there something else we can try or check?

Hi all. I'm having issues logging in to Pupil Cloud

Every time I try to sign in this morning using either the Google login or the direct login, it seems to initially work but I then get redirected to the login screen

I've cleared cache, tried a different browser, and restarted my PC. Not the first time this has happened either, seems to happen every few days

I've just tried a different computer as well, same issue

Hi @user-c1bd23 - thanks for reaching out. We're currently working on fixing this issue! Thanks for your understanding! 🙏🏽

Hi there, I'm encountering an issue where some recordings that I uploaded to pupil cloud is still showing processing (purple loading circle) for the whole day, it usually processes very fast on pupil cloud. I was wondering if there's anyway to fix this issue?

Hi @user-c1bd23 and @user-d2d759 👋. Thanks for your report. We're aware of the issue and are working to resolve it. We'll post here once everything is sorted!

Hi, can the yaw measurement from the IMU be interpreted as cardinal directions? For example, does 0 degrees correspond to north, and so on? I was a bit confused because I initially assumed this was the case, but I’ve seen some discussions on Discord referencing quaternions.

Hi PL team. Same problem again as yesterday. Everything was working fine until about 20 minutes ago, then the fixations stopped appearing in Manual Mapper and I got logged out. Now trying to log back in I get redirected just like yesterday. I managed to log back in successfully once and do some more mapping, but it's logged me out again and I'm unable to sign in

Was the underlying cause to this found? It keeps happening whilst I'm working on mapping, currently trying to meet a project deadline