BiTipText: Bimanual Eyes-Free Text Entry on a Fingertip Keyboard
As computing becomes ubiquitous, the need to provide users with a fast, subtle, and always-available mechanism for text entry has grown significantly. Micro thumb-tip gestures can deliver on this promise, by allowing a user to type by tapping on a miniature keyboard residing invisibly on the first segment of an index finger, using the thumb (TipText). Thus, the text input can be carried out unobtrusively and even without the user looking at the keyboard (referred to as “eyes-free” in this paper). This can lead to better performance when compared with eyes-on input and can also save screen real estate for devices with limited screen space. However, the existing technique is exclusively unimanual, despite typing often being a two-handed activity.
In this paper, we propose a keyboard design for bimanual thumb-tip text input. With two index fingers, the size of the input space doubles, thus the keys are larger and less crowded, which is helpful for reducing tapping errors. Additionally, keys residing on different index fingers use two separate input spaces, thus they can no longer be confused with each other by the system. This largely mitigates the ambiguity issue that is inevitable on a miniature fingertip keyboard. Further, the handedness of the keys on a layout (i.e. which index finger a certain key resides on) determines how typing alternates between the two hands (e.g., left->left->left->right->left). For words with a unique order of handedness, they cannot be typed incorrectly as long as the user types the words in the correct order of handedness, even if the keys are not tapped precisely. All of these benefits can make typing more accurate and faster.
We developed an interactive skin overlay, similar to the one presented in TipText. Our sensor measures ~2.2cm × 2.2cm and contains a printed 3×3 capacitive touch sensor matrix with diamond-shaped electrodes of 5 mm diameter and 6.5mm center-to-center spacing. Our prototype was developed using a flexible printed circuit (FPC). The sensor is 0.025 – 0.125 mm thick and 21.5mm × 27mm wide. The sensor was controlled using an Arduino microcontroller with a MPR121 touch sensing chip. The raw capacitive data from each channel was transmitted at a frequency of 100Hz. Software that interpolates the touch events was implemented in C#.
Selected Press Coverage
Geeky Gadgets: BiTipText finger keyboard uses an interactive skin overlay