Imagine wearing a T-shirt that not only fits comfortably but also tracks your exercise routine with precision. That’s exactly what researchers at Cornell University have achieved with SeamFit, a smart T-shirt that blends seamlessly into your everyday wardrobe while offering advanced exercise tracking capabilities.
Unlike traditional body-tracking garments that can feel bulky and awkward, SeamFit uses flexible conductive threads. These threads are woven into a regular-looking shirt, making it feel just like any other piece of clothing you own. The beauty of SeamFit is that it eliminates the hassle of manual exercise logging. Thanks to its AI capabilities, the shirt can detect your movements, identify different exercises, and even count your reps for you.
What’s more, after your workout, you can simply detach the circuit board from the neckline and toss the shirt into the wash—just like any other T-shirt. Catherine Yu, a doctoral student leading the project, shared her vision: “We wanted to make clothing smart without sacrificing comfort or practicality.”
The SeamFit project, which is detailed in the Proceedings of the ACM on Interactive, Mobile, Wearable, and Ubiquitous Technologies, was put to the test with 15 volunteers. They performed a variety of exercises, including lunges and sit-ups, and the technology scored an impressive 93.4% accuracy in classifying exercises and counting repetitions.
How does it work? SeamFit relies on changes in capacitance within the conductive threads. These changes are transmitted via Bluetooth to a computer for analysis. This innovative approach builds on an earlier project called SeamPose, marking a significant advancement in unobtrusive smart clothing technology. Co-author Cheng Zhang noted the potential of SeamFit to enhance human-AI interaction by enabling AI systems to better understand human activities, thus improving their responsiveness and utility. “We offer a neat form factor that’s always with you,” Zhang explained.
Looking to the future, the research team is exploring scalable manufacturing methods using industrial serger machines. Their goal is to make smart clothing as commonplace as regular clothing. Yu envisions a time when “there’s really no difference between smart and nonsmart clothing.”
