RidgePad: Doubling Touch Accuracy

RidgePad is a new type of touch device that is twice as accurate as traditional capacitive devices, such as those used in Apple's iPhone. Accuracy improvements can be used to (1) make touch UIs more reliable, (2) reduce the need for targeting aids (e.g., shift) & (3), assuming future miniaturization, to bring reliable touch input to substantially smaller mobile devices.

The Underlying Model

RidgePad is based on insights we gained from a series of user studies. We found that precise touch can be achieved if the device senses the full 3D posture of the user’s finger--not just its position on the surface. This is a departure. Since screens are 2D, traditional touch devices take only 2D features of the user’s finger into account when determining touch location. Our findings indicate that this is an oversimplification. We ran a user study that found that error (measured as spread between sensed contact points) is very low for any particular finger angle, but that each angle has a different offset between touch centers and target. We conclude that the spread observed by traditional touch devices does not stem from some inherent inaccuracy of touch (e.g., softness of user’s finger tip) but from the lack of per-posture calibration. Similarly, users are not inaccurate—they are just different, thus require user-specific calibration. We conclude that traditional touch devices are inaccurate, because they are oblivious of 3D finger posture and user ID. The insight is independent of any particular touch device.

The Device

ased on these findings, we have made two highly accurate touch devices that exploit finger posture and user ID in order to increase accuracy. The particular design we have developed, ridgePad, extracts user ID and finger angles from the user's fingerprint. It extract finger angles by locating of key features, such as the "core point" within the fingerprint; a core point shifted to the right, for example, indicates that the finger is tilted to the left. RidgePad doubles touch accuracy with respect to a traditional capacitive pad. A “gold standard” device constructed using an optical tracker shows that the finger posture-based approach can even triple touch accuracy.

Summary of CHI 2010 paper

(a) The Generalized Perceived Input Point Model: a user has repeatedly acquired the shown crosshairs using finger postures ranging from 90° (vertical) to 15° pitch (almost horizontal). The five white ovals each contain 65% of the resulting contact points. The key observation is that the ovals are offset with respect to each other, yet small. We find a similar effect across different levels of finger roll and finger yaw, and across users. We conclude that the inaccuracy of touch (dashed oval) is primarily the result of failure to distinguish between different users and finger postures, rather than the fat finger problem.

(b) The ridges of this fingerprint belong to the front region of a fingertip. Our RidgePad prototype uses this observation to deduce finger posture and user ID during each touch. This allows it to exploit the new model and obtain 1.8 times higher accuracy than capacitive sensing.

Publications

Holz, C. and Baudisch, P. 2010. The Generalized Perceived Input Point Model and How to Double Touch Accuracy by Extracting Fingerprints. In Proceedings of CHI'10, 581–590.

PDF (1.4M) | Presentation (9.8M)