KAIST Logo
CHI 2022 🏅 Honorable Mention Award

Elevate: A Walkable Pin-Array for Large Shape-Changing Terrains

Jaemin Shin , Seungjoo Lee , Taesik Gong , Hyungjun Yoon , Hyunchul Roh , Andrea Bianchi , Sung-Ju Lee

Various automated eating detection wearables have been developed to monitor food intake, addressing the limitations of manual journaling. However, many existing systems lack accuracy in real-world settings or have intrusive designs like headgear. Eyeglasses present a socially acceptable alternative, but current solutions require custom frames and high power consumption. We introduce MyDJ, an eating detection system attachable to any eyeglass frame, which combines a piezoelectric sensor and an accelerometer to capture chewing signals accurately and efficiently. Tested with 30 subjects in uncontrolled environments, including a week-long trial with six users wearing MyDJ on their own glasses, the system achieved a 91.9% F1-score in detecting eating episodes and offered 5.1 times longer battery life compared to existing technologies. Participants also found MyDJ nearly as comfortable (94.8%) as regular eyeglasses.


Reference

@inproceedings{10.1145/3491102.3502041,
  author = {Shin, Jaemin and Lee, Seungjoo and Gong, Taesik and Yoon, Hyungjun and Roh, Hyunchul and Bianchi, Andrea and Lee, Sung-Ju},
  title = {MyDJ: Sensing Food Intakes with an Attachable on Your Eyeglass Frame},
  year = {2022},
  isbn = {9781450391573},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  url = {https://doi.org/10.1145/3491102.3502041},
  doi = {10.1145/3491102.3502041},
  booktitle = {Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems},
  articleno = {341},
  numpages = {17},
  keywords = {automated dietary monitoring, eating detection, multimodal sensing, wearable computing},
  location = {New Orleans, LA, USA},
  series = {CHI '22}
}
Back to projects