top of page

Melting Points

Wearable interactive installation | Gaze Research | Biosensor & hardware | Generative design

Gaze behaviors have hidden information, by just a simple look can reveal the judgment or thoughts from the people. “Melting points” is a series of interactive wearables to track the human gaze about four tasks of judgment and first record the patterns, which are then translated to the wearables made of wax, whose melting behavior corresponds to the data collected from gazing. The project uses computational design, hololens eye tracking system, machine learning data integration, and hardware control system. It is constituted by 4 narratives around human judgment contained within the action of gazing, which explores the concepts of most common social judgment: beauty, wealth, morality, and intellect: what are they, who, or what has the power to define them. The project comes from a study by Yarbus, 1967 [eye movement and vision], “Depending on the task in which a person is engaged, i.e., depending on the character of the information which he must Obtain, the distribution of the points of fixation on an object will vary correspondingly..”

TIME
November 2022

TOOL
Hololens
Arduino
Python
Machine learning
Unity development
Computational design



 

MATERIALS
3D Printing (resin)
Wax
Resistance wire

CONCEPT

research.png

The project comes from a study by Yarbus, 1967 [eye movement and vision] “Depending on the task in which a person is engaged, i.e., depending on the character of the information which he must Obtain, the distribution of the points of fixation on an object will vary correspondingly. because different items of information are usually localized in different parts of an object. ” Yarbu’s research used Repin’s painting “An unexpected visitor” to study gazing behaviors by posing different questions to the gazer. For instance, the researcher asked the gaze to"estimate the material circumstances of the family shown in the picture," in response to which the observer would pay particular attention to the women's clothing and the furniture (the armchair, stool, table chloth, and so on). And following the instruction to “give the ages of the people shown in the picture," the observe would focus on the painted faces.

Coutrot et al. 2016 has a similar study that demonstrated by giving the gaze pattern behaved on human’s portrait, people can predict the object’s gender. The study probe the temporal dynamics of gaze to explore what information can be extracted about the observers and what is being observed. The results show that female gazers follow a much more exploratory scanning strategy than males. Moreover, female gazers watching female actresses look more at the eye on the left side.

These studies illustrate that human gaze behavior contains much invisible information, both observers and what is being observed, like their emotion, attitude, thought, and judgment. In the project, artist chose four most common judgments (beauty, wealth, morality, and intellect) to collect gaze patterns according to these questions. What are they and who or what has the power to define them? No matter whether individual or society, gaze behavior will provide the answer.

 

interaction.png

The project contains two parts, the hololens eye-tracking system with machine learning data analysis, and wax wearable with heating electronic modules inside. With gazers put on the hololens, it will track the eye movement. By looking at the wearable, the wax will melt according to the gaze pattern.

By the gazer standing in front of the people being gazed, with a Hololens 2 on, gaze pattern will record. There are four questions asked and the gazer will use eye movement to answer:

“Is this person appear to be pretty or less pretty?”
“Is this person appear to be rich or poor”
“Is this person appear to be kind or mean?”
“Is this person appear to be smart or less smart”

After collecting the gaze pattern across different individuals, the machine learning will aggregate the gaze behavior and translate to the wearables made of wax.

 

data.gif
making.png
bottom of page