Orbits: Gaze Interaction for Smart Watches using Smooth Pursuit Eye Movements
summary
Orbits is a gaze-only input that places moving targets on a watch face and detects selections by correlating eye and target trajectories with smooth pursuits using a Pearson correlation window of about one second (30 samples at 30 Hz). Evaluation with Pupil Pro and Callista 300 yields 96% correct activations and ~0.73% false positives, scales up to 16 targets before accuracy drops, and powers three working apps: music player, notifications faceplate, and missed call menu.
AI-written summary · about
chapters (11)
- 0:00Start
- 4:58Smooth Pursuits
- 7:14The Correlation Threshold
- 9:34The Orbits Performance on Smartwatches
- 11:19Example Applications
- 11:30Music Player
- 12:24The Notifications Faceplate
- 13:17Contextual Menu
- 13:47Summary
- 15:09Questions
- 17:30Distance Minimum or Maximum Distance
related talks
14mTransform intent into action through your eyes
19mEveryday Eye Contact Detection Using Unsupervised Gaze Target Discovery
42mEyes are the windows to our souls
17mGaze Augmentation in Egocentric Video Improves Awareness of Intention
19mDesigning Multi-Touch Gestures Using the Space around the Smartwatch as Continuous Input Space
10mEyeSwipe: Dwell-free Text Entry Using Gaze Paths
21mHands-Free Coding With Gaze Control in JavaScript
24mGaze and Touch Interaction on Tablets