Abstract: The present disclosure provides systems and methods that leverage machine learning to perform user input motion prediction. In particular, the systems and methods of the present disclosure can include and use a machine-learned motion prediction model that is trained to receive motion data indicative of motion of a user input object and, in response to receipt of the motion data, output predicted future locations of the user input object. The user input object can be a finger of a user or a stylus operated by the user. The motion prediction model can include a deep recurrent neural network.

Abstract: The present disclosure provides systems and methods that leverage machine learning to perform user input motion prediction. In particular, the systems and methods of the present disclosure can include and use a machine-learned motion prediction model that is trained to receive motion data indicative of motion of a user input object and, in response to receipt of the motion data, output predicted future locations of the user input object. The user input object can be a finger of a user or a stylus operated by the user. The motion prediction model can include a deep recurrent neural network.

Abstract: The present disclosure provides systems and methods that leverage machine learning to perform user input motion prediction. In particular, the systems and methods of the present disclosure can include and use a machine-learned motion prediction model that is trained to receive motion data indicative of motion of a user input object and, in response to receipt of the motion data, output predicted future locations of the user input object. The user input object can be a finger of a user or a stylus operated by the user. The motion prediction model can include a deep recurrent neural network.

Abstract: A lighting device includes a substrate, a control circuit, multiple lighting elements and an opaque division plate. The control circuit, the lighting elements and the division plate are mounted on the substrate. The control circuit has a sensor. The lighting elements are electrically connected to the control circuit. The division plate is located between the control circuit and the lighting elements. Given the opaque division plate installed between the lighting elements and the control circuit, light emitted from the lighting elements fails to penetrate through the division plate and is thus not sensed by the sensor on the control circuit opposite to the lighting elements. Accordingly, when detecting a dark environmental luminance, the control circuit activates the lighting elements, and after the lighting elements are activated, light emitted from the lighting elements won't cast on the sensor to affect correct determination of the control circuit.

Abstract: The present disclosure provides systems and methods that leverage machine learning to perform user input motion prediction. In particular, the systems and methods of the present disclosure can include and use a machine-learned motion prediction model that is trained to receive motion data indicative of motion of a user input object and, in response to receipt of the motion data, output predicted future locations of the user input object. The user input object can be a finger of a user or a stylus operated by the user. The motion prediction model can include a deep recurrent neural network.

Abstract: A method of messaging with light is implemented by a light fixture. The method includes: timing and recording a switch-on period, and recording an operation mode of the light fixture during receipt of the electrical power; and operating in a messaging mode the next time the light fixture receives electrical power upon determining that the recorded operation mode is a lighting mode and that the recorded switch-on period is within a predetermined range of time period. In the messaging mode, the light fixture controls a light emitting device to send a binary message by operating in a sequence composed at least of a first brightness state and/or a second brightness state.

Abstract: Alight fixture receives AC power via a light switch to drive light emission of a light emitting module, and is switchable among multiple states that include a brightness selecting state and a selected brightness state. The light fixture causes the light emitting module to cyclically emit light according to multiple brightness levels in sequence in the brightness selecting state, and to emit light according to one of the multiple brightness levels in the selected brightness state.

Abstract: A cosmetic composition containing fragments of bacterial cellulose film is provided. The addition of the fragments of bacterial cellulose film in a cosmetic composition not only improves the transdermal transmission of active ingredients contained in the cosmetic composition but also provides the functions of skin moisturization, skin exfoliation and sebum absorption. Various applications of bacterial cellulose film are thus provided.