VergeIO: Depth-Aware Eye Interaction on Glasses

TL;DR

VergeIO introduces a depth-aware eye interaction system for glasses by sensing vergence with a two-channel EOG electrode layout optimized for visibility and comfort. It supports six vergence gestures across three depths with personalized accuracies of 83–98% and cross-user generalization of 77–97% without calibration, while consuming ~3 mW and maintaining 15 ms latency. A comprehensive artifact-removal pipeline, a preamble activation scheme, and compatibility with traditional EOG signals drive robust real-world performance; virtual and approximate depth cue generalization expand applicability. The work presents a compact hardware prototype, an end-to-end processing pipeline, and open-source plans, enabling new consumer and healthcare depth-aware eyewear applications.

Abstract

There is growing industry interest in creating unobtrusive designs for electrooculography (EOG) sensing of eye gestures on glasses (e.g. JINS MEME and Apple eyewear). We present VergeIO, the first EOG-based glasses that enables depth-aware eye interaction using vergence with an optimized electrode layout and novel smart glass prototype. It can distinguish between four and six depth-based eye gestures with 83-98% accuracy using personalized models in a user study across 20 users and 2,400 gesture instances. It generalizes to unseen users with an accuracy of 77-97% without any calibration. To reduce false detections, we incorporate a motion artifact detection pipeline and a preamble-based activation scheme. The system uses dry sensors without any adhesives or gel, and operates in real time with 3 mW power consumption by the sensing front-end, making it suitable for always-on sensing.

Figures (19)

• Figure 1: Applications of VergeIO. (a) Varifocal lenses adjust focal depth as the user shifts gaze between ambient reference points: a nearby phone and a distant TV. (b) Smart glasses enable gaze-based selection between virtual objects at different depths. (c) Remote screening of ocular disorders using approximate reference points: thumbs and a distant object.
• Figure 2: Geometric characterization and signal validation for vergence distance selection.(a)VergeIO measures changes in vergence angle ($\theta$) using EOG signals, which vary proportionally with these changes. (b) Our system can distinguish changes in vergence angles corresponding to three interaction distances: 30, 70, and 200 cm. (c) Mean EOG signal for each of the six vergence gestures for the left and right eye. Shaded region represents one standard deviation from the mean.
• Figure 3: Limitations of prior EOG electrode placements on glasses. (a) Design of JINS MEME jinsmeme and $\text{GAP}{\text{ses}}$frey2024gapses lacks spatial separation which hinders vergence detection. Source: https://www.ohmyglasses.jp/blog/2016/05/31/jins-meme-new-world/(b)VergeIO places electrodes at the temples and nose bridge which increases sensitivity of vergence detection. By introducing a vertical separation, the design is compatible with traditional horizontal and vertical EOG sensing. Flexible contact arms for the electrodes ensure stable skin contact across users while being compatible with a glasses form factor.
• Figure 4: SNR comparison across six vergence gestures.VergeIO outperforms the commercial JINS MEME glasses jinsmeme.
• Figure 5: Hardware design. (a) Exploded CAD design illustrating the different components of the hardware design. (b) Fabricated design annotated with location of different components. (c) Front-view of glasses with dimensions.