.png){kind=icon}
.png){kind=icon}
Return
PalmPen: A Palmprint Positioning Pen
.png){kind=icon}
PRE
AI
J
王
Z
冯
周
DOI：10.1145/3749549.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Palm-based interaction offers unique advantages in tasks requiring minimal visual attention and low cognitive load, thanks to the sensitive tactile feedback of the palm and the body's proprioceptive abilities. However, current research in this field faces challenges like limited detection precision and complex equipment. This paper presents PalmPen, a novel pen-shaped camera device that relies on computer vision algorithms to enable absolute positioning and continuous tracking on the palm. We created a training dataset containing full and partial palmprints with corresponding positional data using cameras and an optical tracker. A deep learning network was created to predict the location of partial palmprints within a full palmprint. PalmPen achieved a mean positioning error of 2.74 mm in the experiments. A user study with 12 participants demonstrated that PalmPen offers superior positioning accuracy and input efficiency over other methods. Additionally, properties such as touch pressure and rotation angles can be inferred from palmprint sequences, extending PalmPen's application range.
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
P
IF:
4.5
Papers:
1.1K
Citations:
7.2K
Organization
No organization information available
Cited Papers