Intel Corporation

Korea Advanced Institute of Science and Technology

Jinwon Lee

Giuseppe Beppe Raffa

Lama Nachman

Jonathan J Huang

Rahul C Shah

Hong-Young Chang

Sanghyuk An

Embodiments of the invention describe a system to efficiently execute gesture recognition algorithms. Embodiments of the invention describe a power efficient staged gesture recognition pipeline including multimodal interaction detection, context based optimized recognition, and context based optimized training and continuous learning. Embodiments of the invention further describe a system to accommodate many types of algorithms depending on the type of gesture that is needed in any particular situation. Examples of recognition algorithms include but are not limited to, HMM for complex dynamic gestures (e.g. write a number in the air), Decision Trees (DT) for static poses, peak detection for coarse shake/whack gestures or inertial methods (INS) for pitch/roll detection.

Efficient gesture processing

A plasma generation apparatus includes a vacuum container, dielectrics connected to through-holes formed in the vacuum container, RF coils of the same structure disposed in the vicinity of the respective dielectrics and electrically connected in parallel, an RF power source to supply power to the RF coils, an impedance matching circuit disposed between the RF power source and the RF coils, and a power distribution unit disposed between the impedance matching circuit and one ends of the RF coils to distribute the power of the RF power source to the RF coils. The power distribution unit includes a power distribution line and a conductive outer cover enclosing the power distribution line. Distance between an input end of the power distribution unit and the RF coils are equal to each other, and the other ends of the RF coils are connected to the conductive outer cover to be grounded.

Plasma generation apparatus

Systems and methods for proximity detection between electronic devices are disclosed. One or more electronic devices transmit signals to a proximity server, which determines whether the first electronic device may be proximate the second electronic device. The proximity server transmits a signal to the first electronic device and the second device, and in response to the signal, the first and second electronic devices activate an environmental sensor, collect at least one sample of environmental data, extract at least one feature set of the environmental data, generate a first obscured feature from the feature set, transmit the first and second obscured feature sets to the proximity server. The proximity server uses the first obscured feature set and the second obscured feature set to determine whether the first electronic device and the second electronic device are proximate.

Multimodal proximity detection

Growing community of inventors

Daejeon, South Korea

Jinwon Lee