Skip to main content

using machine learning to detect and improve engagement

Project description

vroom

(yes, the name is kind of dumb; deal with it)

There are a lot of sections so scroll through and read the headers to tell which information will be useful to you.

Teachers

Here's the part especially relevant for teachers reading this: Let's take smiling, which has been shown to improve happiness levels. You can integrate this program into a lesson. For example, ask your students to smile to show they've understood what you've said. Now you're asking everyone to interact with the lesson in a way that has a positive impact and is likely to increase their interest in the lesson. In the future, different hand gestures and more expressions will be added, expanding the range of gestures you can use to improve their engagement.

Students

So you're a student, who, for whatever reason, doesn't want to turn on their camera. Teachers aren't legally allowed to make you do so, but that doesn't seem to stop them (some of them at least – #notallteachers). Wouldn't it be cool to be able to show the teacher that you're still paying attention without showing your face? (Yes, it would, is the correct answer. No other options allowed.) (Arguably) even cooler, you can raise your hand without having to press that weird yellow hand reaction in zoom, which isn't terrible – it's the wrong skin color no matter who you are and I respect that – actually no it's pretty terrible that's the only good thing I have to say about it.

Ever spent hours at a time/an entire afternoon obsessively creating picrews? (iykyk) Well, now's the time to put them to good use. Sure, regular profile pictures are fun, but what about a profile picture that reflects your actions? If you squint, it even looks like animation.

If you don't know what picrews are, they're basically customizable avatars. When you downlaod this program, it comes with a very generic one as an example, but the whole point is for you to change it!! Here are some that you can try:

https://picrew.me/image_maker/420013

https://picrew.me/image_maker/344854

https://picrew.me/image_maker/567717

https://picrew.me/image_maker/137904

https://picrew.me/image_maker/154823 (one of my favorites)*

https://picrew.me/image_maker/14314 (another favorite)*

I don't like these ones as much but they are probably still useful for others: https://picrew.me/image_maker/407027

https://picrew.me/image_maker/644129

https://picrew.me/image_maker/565738

*Yes, it is more useful if the images match up with the expressions (squint, wide-eyed, smile -- more to come!!), but these are the best anyways.

Side note: I sincerely doubt that anyone will be judging your picrews – I wouldn't and I'm a very judgemental person – but, hey, I get it. You can still use it with friends for fun (anyone have online friends they're not allowed to show their face to? no? just me?) and obviously, there's no restriction on which images you choose. (Yes, that does mean you can use memes, raccoon pictures, etc. Who can stop you?). Why use this instead of just setting your profile picture to your picture of choice? Well, consider this: not one raccoon/meme/raccoon meme, but four.

Also featuring a rough graph of engagement during a call that students can view, just for funzies (Am I using that ironically? Yes. Am I aware that using a term or phrase ironically usually leads to using it unironically? Also yes, but we're going to ignore that for now.)

Another IMPORTANT side note: Currently, this program works by changing your profile picture in real time, so you have to turn off your video to use it, and it only works on Zoom. But, in the future, look out for a version that creates an entirely new stream for you to select as your video source in any video call application. Also, that doesn't make you run the program in terminal every time you want to use it. Basically just an overall better version.

Ok, pitch over. Want to learn about this project without suffering through my attempts to be funny? Just keep reading, past the Instructions section (although that's probably also useful) to the second star on the right (i.e., the project summary section).

Instructions

Download:

(Don't move the downloaded folder from Downloads. If you really have to, it's easy to fix the code if you already have Python experience. You can also contact me and I can send you a version of the program that works for you.)

Change the Images: In the downloaded folder, you'll find four images: defaultpic.png, smilepic.png, squintpic.png, and widepic.png. Hopefully the names are pretty self-explanatory. When you have your own versions of a default picture, a smiling picture, a squinting picture, and a wide-eyed picture, delete the existing ones and drag yours into the folder. Make sure to name them the exact same things.

Run:

After: Access your reports in the same folder that you downloaded. There should be two files titled with the date – a .csv file and a .png. Please take 30 seconds to fill out the survey (no hard questions, I promise): https://docs.google.com/forms/d/e/1FAIpQLSe3DtjxwRutiEEWN05C-BuC_I97shJp9bf0yORIDca1_6NWkA/viewform?usp=sf_link

Project Summary This project will:

  • Help teachers invite students to engage more with online teaching.
  • Help students interact more with the lesson while still being able to turn their camera off.
  • Help students analyze their own engagement in the lectures they are attending.

Students can download an app that will

  • Analyze the live webcam video in real time.
  • Display the aforementioned actions/emotions during video calls by replacing the live webcam with a user-selected image, communicating a specific action/emotional indicator.
  • Return a report to the student with a breakdown of their engagement with the lecture.

Users need to start the utility on their laptop while they are attending a lecture. Once the lecture is done or earlier if they wish, they can stop analysis.

Install by downloading zip file, as well as all dependencies in requirements.txt

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

vroompkg-0.0.7.tar.gz (82.5 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

vroompkg-0.0.7-py2.py3-none-any.whl (82.9 MB view details)

Uploaded Python 2Python 3

File details

Details for the file vroompkg-0.0.7.tar.gz.

File metadata

  • Download URL: vroompkg-0.0.7.tar.gz
  • Upload date:
  • Size: 82.5 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.3.0 pkginfo/1.7.0 requests/2.25.1 setuptools/53.0.0 requests-toolbelt/0.9.1 tqdm/4.57.0 CPython/3.8.2

File hashes

Hashes for vroompkg-0.0.7.tar.gz
Algorithm Hash digest
SHA256 5896d9701936da9a1c3213f8a8fc1a8dc3f52a8d3ad0249bbe5fd8960702b96c
MD5 4b25ce3d8bf00975998bf559f1a5e617
BLAKE2b-256 ac4f1bc4400f6888bf2104ccc6cf7b213429ce1c83d0ddbdd30aea8d0cce9d90

See more details on using hashes here.

File details

Details for the file vroompkg-0.0.7-py2.py3-none-any.whl.

File metadata

  • Download URL: vroompkg-0.0.7-py2.py3-none-any.whl
  • Upload date:
  • Size: 82.9 MB
  • Tags: Python 2, Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.3.0 pkginfo/1.7.0 requests/2.25.1 setuptools/53.0.0 requests-toolbelt/0.9.1 tqdm/4.57.0 CPython/3.8.2

File hashes

Hashes for vroompkg-0.0.7-py2.py3-none-any.whl
Algorithm Hash digest
SHA256 25f2fea80350a67f8673b545a55efcefbb39fa40a43f1f3eb8dd30ce952fe688
MD5 c12b54c45448229667c1fe9642914146
BLAKE2b-256 cbe66f1d29cbf590c639046e5728041b3205e2e6a99623ccf8ef28bbe17409b3

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page