Finger Lakes Today

Cornell researchers have developed a wristband device called EchoWrist that uses AI-powered, inaudible soundwaves to continuously detect hand positioning and hand-object interactions. Cheng Zhang, assistant professor of information science at Cornell, emphasized the importance of hands in daily activities, stating, “The hand is fundamentally important – whatever you do almost always involves hands.”

Chi-Jung Lee, one of the doctoral students involved in the project, highlighted the potential applications of EchoWrist, including tracking hand positions for virtual reality systems and enabling one-handed interactions with devices like smartphones. The device, small enough to fit onto a commercial smartwatch, was developed by researchers at the Cornell Ann S. Bowers College of Computing and Information Science.

EchoWrist, presented at the Association of Computing Machinery CHI conference on Human Factors in Computing Systems (CHI’24), also allows users to control devices with gestures and give presentations. Lee mentioned, “We can enrich our interaction with a smartwatch or even other devices by allowing one-handed interaction – we could also remotely control our smartphone.”

The innovative technology behind EchoWrist involves two tiny speakers and microphones mounted on a wristband, enabling the device to bounce inaudible sound off the hand and objects for accurate tracking. Ruidong Zhang, another doctoral student involved in the project, highlighted that EchoWrist not only tracks the hand but also the surrounding environment.

In testing, EchoWrist demonstrated a 97.6% accuracy in detecting objects and actions, making it suitable for interactive applications like following recipes without getting screens dirty. The device's usage of acoustic tracking enhances user privacy while maintaining high performance levels.

Cheng Zhang expressed excitement about the potential applications of EchoWrist, stating, “One of the most exciting applications this technology would enable is to allow AI to understand human activities by tracking and interpreting the hand poses in everyday activities.”

Despite some limitations in distinguishing between objects with similar shapes, such as a fork and a spoon, researchers are confident in the technology's potential for improvement. With further optimization, EchoWrist could easily be integrated into existing smartwatch technology.

The project, funded by the National Science Foundation, involved a team of researchers, doctoral students, and undergraduates from various institutions. EchoWrist represents a significant advancement in low-power, body pose-tracking technology, offering promising possibilities for the future of human-computer interactions.