M.Gesture: An Acceleration-Based Gesture Authoring System on Multiple Handheld and Wearable Devices
summary
M.Gesture is a smartphone/watch authoring app that visualizes accelerometer streams with a mass-spring metaphor and lets users define gestures by demonstration plus graphical declaration using intersecting hurdle planes. Built for single or multi-device setups, it runs a hurdle filter then DTW on candidates, yielding about 88% precision, ~4 percentage points better than plain DTW and 26% faster. Lab studies (20 participants, ~1,000 gesture trials) show average composition under 100 seconds and 90% under 2.5 minutes; rotational visibility and long‑term robustness remain open issues.
AI-written summary · about
chapters (5)
- 0:00Introduction and motivation
- 3:59M.Gesture system design
- 8:29User study results
- 13:33Limitations and future work
- 14:38Conclusion and Q&A
related talks
06mDesigning and Evaluating Hand-to-Hand Gestures with Dual Commodity Wrist-Worn Devices
24mFrom Touch To Code: Gestures and Beyond
19mDesigning Multi-Touch Gestures Using the Space around the Smartwatch as Continuous Input Space
19mBackHand: Sensing Hand Gestures via Back of the Hand
17mUser-Driven Design Principles for Gesture Representations
20mExpressive Keyboards: Enriching Gesture-Typing on Mobile Devices
43mDesigning for Hands, Gestures, and VR
16mGesturePod: Enabling On-device Gesture-based Interaction for White Cane Users