Eye movements have been widely studied in vision research, language and usability, yet progress has been limited since eye trackers are expensive (>$10K) and do not scale. In this talk, we'll present findings from our recent paper at Nature communications, which shows that smartphone's selfie cameras+ML can achieve high gaze accuracy comparable to SOTA eye trackers that are 100x more expensive. We demonstrate that this smartphone technology can help replicate key findings from prior eye movement research in Neuroscience/Psychology, that earlier required bulky/expensive desktop eye trackers in highly controlled settings (e.g., chin rest).

These findings offer the potential for orders-of-magnitude scaling of basic eye-movement research in Neuroscience/Psychology (with explicit user consent) and unlock new applications for improved accessibility, usability and screening of health conditions