User Experience Researcher


We designed a way to manipulate and deceive the Zoom facial recognition algorithm.


Design a way to manipulate and deceive perceptual cognition


Munira Kazi, Rebecca Hodge and Devin Wang 

Methods used

Artefact analysis, Information visualisation


24.11 – 09.11.2022


Based on the feedback we have received we decided to focus on disguising the facial emotions in a context where only the bust is visible.

This finding connected well with our visual research from artists who done masks and attempted to distort the human face through cutouts and wearables that use optical illusions.

We started experimenting with what we have learned to understand it better and see where we can push it to.

Cutouts from the magazine were good for someone that is static however there were issues with attachment and the solid nature of the cutouts which prevented vision or talking making them very inconvenient.

The nylon material felt very artificial where in cutouts case felt like a beauty filter whereas in the case of a whole mask it was also impractical for the wearer’s perception and comfort: it was hard to see once the paint was applied on top of it. The rice paper mask had similar qualities and even more seemed like we are moving to concealment than disguising. 

Cutouts prototype
Nylon mask prototype

The wired mesh material worked well in terms of comfort and stayed in shape but it was complicated to manipulate the flat surface of it to follow the intricate shape of the human shape. There were easily solvable yet noticeable issues with the shininess of the material and hazards around the cutout material. Fabric mesh was a more accessible solution but the problem was that it could not hold shape at all unless introduced other materials.

Drawing the wired mesh face mask
Cutting the wired mesh face mask
Wearing the wired mesh face mask. Photo credit: Rebecca Hodge
Wearing the fabric mesh face mask. Photo credit: Rebecca Hodge

Through our personal experiences with online tools for communication, we noticed a real-world context for disguising privacy and other problems related to their use.

New direction:

How might we disguise ourselves from the online Zoom calls while being present at them? 

Reducing the field of view of webcam. Photo credit: Munira Kazi
Trying red filter over the webcam. Photo credit: Munira Kazi

Experimenting with facial recognition – Google Teachable Machine

This website allowed us to perceive like a machine which immensely helped us. Our experiments with the AI yielded the result when we covered just enough of the face above and below the eyes, which was contrary to the hypothesis we had.

With our discovery, we started designing wearables playing to scope the limits of machine recognition and also make it comfortable for the human wearer.

First wearable prototype successfully confusing the facial recognition algorithm
Second iteration of the wearable prototype
Third version of the wearable face mask confusing the algorithm in practice. Photo credit: Devin Wang
Wearable prototype for changing the overall head shape
Another wearable prototype for changing the overall head shape
Pink mesh wearable prototype
Testing the pink mesh wearable prototype
Testing nylon wearable prototype
Testing wearable face mask prototype with a fake human face
Wearable face mask prototype with a fake human face
Wearable prototype with sticks and label “I CAN HEAR YOU” successfully working
User testing prototype with a colleague

Next steps

We could have considered more ergonomics for the wearer – an area we did not have much time to spend.

In the next iteration, we should consider various degrees of privacy for the wearer to choose how much of them to be disguised.


Daily Mail Reporter. (2013). Can you read this face? Artists distort faces in magazine pictures to create bizarre expressions. Daily Mail. Available at: (Accessed: 4 December 2022)

Lynch, E.D.W. (2014). Optical Helmets that Bizarrely Distort the Wearer’s Face. Laughing Squid. Available at: (Accessed: 4 December 2022)

If you liked it share!