HANOVER, N.H. — Smartphones and smart TVs are a common sight nowadays, but what about smart tablecloths? Sounds silly at first, but a team at Dartmouth College have created a smart fabric capable of figuring out what you’re putting on it.
This innovative fabric, known as Capacitivo, uses subtle shifts in electrical charges to identify objects of various shapes and sizes. This includes non-metallic objects like fruits and credit cards.
“This research has the potential to change the way people interact with computing through everyday soft objects made of fabrics,” says senior study researcher Xing-Dong Yang, an assistant professor of computer science, in a university release.
Current fabric sensing techniques currently require some form of input, such as a user’s touch. Capacitivo sets itself apart by using an “implicit input” technique that doesn’t require any action from scanned objects.
How much does the smart tablecloth know?
Electrical charge information is recorded by the tablecloth’s electrodes and then compared to pre-loaded data within the system via machine learning. The fact that the tablecloth can recognize non-metallic goods such as foods, paper, liquids, and plastic is very noteworthy.
“Being able to sense non-metallic objects is a breakthrough for smart fabrics because it allows users to interact with a wide variety of everyday items in entirely new ways,” explains lead study author Te-Yen Wu, a PhD student at Dartmouth.
Study authors performed tests on 20 different items, with the smart tablecloth recording a 94.5 percent accuracy level. More specifically, it is quite adept at identifying different types of fruit, as well as the fullness of liquid containers. It can also differentiate between different beverages like water and soda. It is somewhat less accurate however with lighter objects like plastic credit cards.
To explain Capacitivo’s broad skillset, let’s say your potted plant needs some water. The tablecloth can pick up on that and initiate the appropriate action. This technology can also help with cooking by making recipe recommendations, or help find lost goods and provide alerts.
This research was presented at the ACM Symposium on User Interface Software and Technology (UIST 2020).