Abstract: Disclosed include a data collection device, a data collection system, and an electronic image stabilization device. The data collection device includes a logic circuit. The logic circuit includes: a timer, configured to generate timestamp data; a control module, configured to, in response to receiving a pre-set input signal, generate a data obtaining command and obtain the timestamp data from the timer; a communication module, configured to, in response to the data obtaining command, obtain MEMS (MEMS) data from a MEMS sensor; and a packing module, configured to pack the MEMS data and the timestamp data into a data packet.

Abstract: A control method and apparatus for a new energy camera device, a camera device, a storage medium, and a program product are provided. The control method includes: determining a current electric quantity of the battery assembly and a predicted duration for full charging the battery assembly by the new energy charging assembly; determining an operation mode to be adopted from N preset operation modes of the new energy camera device based on the current electric quantity and the predicted duration for full charging the battery assembly, wherein N is a positive integer of at least 2; and controlling the new energy camera device to operate in the determined operation mode. In the embodiments of the present application, the operation mode to be adopted is determined from a plurality of operation modes based on the current electric quantity and the predicted duration for full charging the battery assembly, which helps to solve the defect of the continuous operation in a low power consumption mode in the related art.

Abstract: Provided in the embodiments of the present application are a brightness measurement method and apparatus, a control method and apparatus for a photographic apparatus, and a medium. In the brightness measurement method, a brightness measurement circuit is used to perform brightness measurement, the brightness measurement circuit comprising: a photosensitive diode, a switch and at least two measurement groups.

Abstract: Embodiments of the present disclosure provide a voice wake-up method, apparatus, electronic device, and readable storage medium. The method includes: acquiring a deep feature encoded by a voice activity detection (VAD) encoding layer through a VAD decoding layer, the deep feature being extracted from a digital audio signal used for VAD encoding and voice wake-up encoding; determining whether each audio segment in the digital audio signal is a voice according to a probability that each audio frame in the digital audio signal is a voice frame, each audio segment including a plurality of audio frames; and when determining that the audio segment is the voice, indicating, by the VAD decoding layer, to input a deep feature of the audio segment into a voice wake-up decoding layer to perform voice wake-up. The embodiments of the present disclosure improve the accuracy of voice wake-up.

Abstract: Embodiments of this application provide a pan-tilt, a method for controlling a pan-tilt, and a photographing apparatus. A magnetic-field generating apparatus for generating uniform detection magnetic field is mounted on a pan-tilt base of the pan-tilt, a conductor rod is mounted on a pan-tilt head of the pan-tilt, and two ends of the conductor rod are electrically connected to a voltage detection unit. In this way, in a process of the pan-tilt head rotating relative to the pan-tilt base, the conductor rod cuts magnetic induction lines of the detection magnetic field and generates an induced electromotive force. In addition, in the embodiments of this application, the induced electromotive force on the two ends of the conductor rod can be detected through the voltage detection unit.

Abstract: The present application discloses a method, apparatus, device, medium and convolutional neural network for quantization aware training. The method is performed by electronic equipment and includes: performing sample training on a first convolutional neural network, wherein at least one convolution layer of the first convolutional neural network includes a mergeable branch structure and at least one first shortcut; reserving the at least one first shortcut in the at least one convolution layer of trained first convolutional neural network; merging the mergeable branch structure except the at least one first shortcut in the at least one convolution layer, and obtaining a second convolutional neural network having a first shortcut structure; and, performing quantization aware training based on the second convolutional neural network to improve accuracy of quantization of the convolutional neural network model.

Abstract: The present application discloses a method for reducing power consumption of a communication device in a sleep state. The method includes: when the communication device is in the sleep state, detecting an running state of a current network, and setting a next keep-alive probing period of a persistent connection according to the running state of the current network such that the next keep-alive probing period of the persistent connection is at least extended on the basis of a current keep-alive probing period when the running state of the current network is an idle state.

Abstract: Embodiments of the present disclosure provide a voice wake-up method, apparatus, electronic device, and readable storage medium. The method includes: acquiring a deep feature encoded by a voice activity detection (VAD) encoding layer through a VAD decoding layer, the deep feature being extracted from a digital audio signal used for VAD encoding and voice wake-up encoding; determining whether each audio segment in the digital audio signal is a voice according to a probability that each audio frame in the digital audio signal is a voice frame, each audio segment including a plurality of audio frames; and when determining that the audio segment is the voice, indicating, by the VAD decoding layer, to input a deep feature of the audio segment into a voice wake-up decoding layer to perform voice wake-up. The embodiments of the present disclosure improve the accuracy of voice wake-up.

Abstract: A memory management method for a device, and a memory management device and a computing system.

Abstract: The present disclosure relates to a training method for a parameter configuration model, a parameter configuration method, and a parameter configuration device.

Abstract: A memory management method for a device, a memory management device, and a computing system.

Abstract: The present disclosure relates to a training method for a parameter configuration model, a parameter configuration method, and a parameter configuration device.

Abstract: A simple solution processing method is developed to achieve uniform and scalable stabilized lithium metal powder coating on Li-ion negative electrode. A solvent and binder system for stabilized lithium metal powder coating is developed, including the selection of solvent, polymer binder and enhancement of polymer concentration. The enhanced binder solution is 1% concentration of polymer binder in xylene, and the polymer binder is chosen as the mixture of poly(styrene-co-butadiene) rubber (SBR) and polystyrene (PS). Long-sustained, uniformly dispersed stabilized lithium metal powder suspension can be achieved with the enhanced binder solution. A uniform stabilized lithium metal powder coating can be achieved with simple doctor blade coating method and the resulting stabilized lithium metal powder coating can firmly glued on the anode surface.

Abstract: The present invention provides for a composition of matter comprising: poly(9,9-dioctylfluorene-co-fluorenone-co-methylbenzoic ester)(PFM), carbon nanotubes (CNT), and sulfur particles nanocomposite, wherein the nanocomposite is porous. The present invention also provides for an electrode comprising: poly(9,9-dioctylfluorene-co-fluorenone-co-methylbenzoic ester)(PFM), carbon nanotubes (CNT), and sulfur particles nanocomposite, wherein the nanocomposite is porous. The present invention also provides for a lithium sulfur (Li—S) battery comprising: an electrode comprising poly(9,9-dioctylfluorene-co-fluorenone-co-methylbenzoic ester)(PFM), carbon nanotubes (CNT), and sulfur particles nanocomposite, wherein the nanocomposite is porous.

Abstract: A simple solution processing method is developed to achieve uniform and scalable stabilized lithium metal powder coating on Li-ion negative electrode. A solvent and binder system for stabilized lithium metal powder coating is developed, including the selection of solvent, polymer binder and enhancement of polymer concentration. The enhanced binder solution is 1% concentration of polymer binder in xylene, and the polymer binder is chosen as the mixture of poly(styrene-co-butadiene) rubber (SBR) and polystyrene (PS). Long-sustained, uniformly dispersed stabilized lithium metal powder suspension can be achieved with the enhanced binder solution. A uniform stabilized lithium metal powder coating can be achieved with simple doctor blade coating method and the resulting stabilized lithium metal powder coating can firmly glued on the anode surface.

Abstract: Techniques for adaptively accessing image data in a data memory space are disclosed. According to one aspect of the present invention, a data transfer bandwidth between a display controller and a memory device is detected to obtain bandwidth characteristics. An image format to be used is determined by comparing the bandwidth characteristics with a preset mapping relationship between at least two image formats and the bandwidth characteristics. A current image format is updated by the determined image format to be used if the determined image format to be used is different from the current image format. The image data in the memory is then read out according to the current image format.