CrossGR: Accurate and Low-cost Cross-target Gesture Recognition Using Wi-Fi
summary
CrossGR claims cross-target Wi‑Fi gesture recognition using a generative adversarial network to synthesize training CSI data from a few users and a CNN-based pipeline that learns user-agnostic features with a feature‑elimination path and a target‑adaptive classifier. The system is implemented on CSI, tested on 15 gestures across 60 volunteers in three rooms, and reports average accuracies of 87.4% (10 gestures) and 82.6% (15 gestures), reducing required real samples by an order of magnitude.
AI-written summary · about
related talks
20mInteracting with Soli: Exploring Fine-Grained Dynamic Gesture Recognition in the Radio-Frequency ...
19mA Rapid Prototyping Approach to Synthetic Data Generation For Improved 2D Gesture Recognition
20mAdvancing Hand Gesture Recognition with High Resolution Electrical Impedance Tomography
19mBackHand: Sensing Hand Gestures via Back of the Hand
28mStrike a Pose - Gesture recognition in JavaScript w/ ML & Arduino
10mA large, crowdsourced evaluation of gesture generation systems on common data: The GENEA ...
19mGroup Touch: Distinguishing Tabletop Users in Group Settings via Statistical Modeling of Touch Pairs
14mBack-Hand-Pose: 3D Hand Pose Estimation for a Wrist-worn Camera via Dorsum Deformation Network
