What is All This About?

A valid question ... we think we're finding out. Many things are broken on the internet of profiling? The commodities we enjoy come hand in hand with a surrender of self. This is general knowledge, commonly accepted, impossible to parry?

Below we list some of our reasoning for making Wealgo. The fun aspect dominates them all.

To Play or to Profile ...

Algoritms - as humans? - are trained to profile. This fixation of inherently fluid concepts such as: time, identity, ethnicity, gender, religion, clearly presents thé predicament our time is currently facing (wordplay intended). WeAlgo aims to provide a space to play not to profile. By decoding the visual language of human ánd non-human, both are lost in translation and 68 facial landmarks a canvas to play with. Come and dance with other false-positives! We do not collect, store, share or sell data.

Private Space

Face recognition is based on the detection of meaningful biological facial landmarks and the extraction of this geometry into a set of - in our case 68 - predefined (x,y)-coordinates for: mouth, eyebrows, eyes, nose and jaw. This technology - ubiquitously used in AR-, law enforcement or entertainment applications - is the basis for your customisable avatar, the vehicle that allows us to be personally present, without surrendering our screen privacy. It both safeguards our private live whilst it makes transparent the information our effigies implicitly provide.

You can be totally anonimous in Wealgo.

Cut Pollution

Instead of distributing compressed video streams, WeAlgo streams data points, which significantly reduces bandwidth, costs and thus carbon footprint. HD video conferencing bandwidth requirements run between 2-3 Mbps(up/down)* per user. Multiply this by the average time spent on a call (1hr) to end up with an average total of 810MB data per user per call.

Wealgo uses client side scripting to compute landmark data points and colors to broadcast a WebRTC data-stream to your Room's participants. For the datapoints we use 31Kbps (up/down) per user. Wealgo uses another 50Kbps (up/down for audio); together making a total data transfer for a similair call (1hr) of 36MB ... or 4.5%.

95.5% savings ... We think make an important difference.

*source: https://support.zoom.us/hc/en-us/articles/201362023-System-Requirements-for-PC-Mac-and-Linux

Customised Alter Algos by Children

Impressions of Alter Algos, created by children.

Open Source, Private, Secure

Wealgo uses WebRTC technology. With WebRTC, you can add real-time communication capabilities to your application that works on top of an open standard. It supports video, voice, and generic data to be sent between peers, allowing developers to build powerful voice- and video-communication solutions. The technology is available on all modern browsers as well as on native clients for all major platforms.

We also experiment with BRFv5 a very smooth client based face recognition algorithm by tastenkunst.com, thank you for offering us a free period license to work with your tools.

The software we develop is available on GitLab and free for anyone to use. We don’t collect or sell data and the IP of the Alter Algo you create is yours.

Team

WeAlgo is a project by Martine Stig, Stef Kolman, Tomas van der Wansem and Roel Noback.