Abstract
The interactions of a user with a smart television may rely on the use of physical remote controls (e.g., remote control devices). Physical remote controls may present challenges related to battery maintenance, frequent misplacement, and limited accessibility. The disclosed technology provides a user interface navigation layer for a television application running on the smart television that provides a remote-free interface to the television application that enables complex spatial navigation on the smart television using edge artificial intelligence (edge AI) hand gesture recognition by way of an integrated or auxiliary camera. The remote-free interface may map specific hand gestures, such as swipes and pinches, to directional intent, acting as a virtual directional pad. To ensure accuracy and low latency, the remote-free interface utilizes on-device edge inference for hand landmarking, temporal smoothing to filter out unintentional hand jitters, and a transformer-based intent classification model to distinguish deliberate commands from casual movements. The disclosed technology may replace a physical remote control by providing fluid, system-level navigation for complex user interfaces and nested menus through intuitive hand motions
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Gupta, Richa, "Architectural Framework for Gesture-Controlled Smart Television Interfaces", Technical Disclosure Commons, (May 17, 2026)
https://www.tdcommons.org/dpubs_series/10140