617results about How to "Enter exactly" patented technology

An apparatus for providing a language based interactive multimedia system includes a selection element, a comparison element and a processing element. The selection element may be configured to select a phoneme graph based on a type of speech processing associated with an input sequence of phonemes. The comparison element may be configured to compare the input sequence of phonemes to the selected phoneme graph. The processing element may be in communication with the comparison element and configured to process the input sequence of phonemes based on the comparison.

A speech recognition method, medium, and system. The method includes detecting an energy change of each frame making up signals including speech and non-speech signals, and identifying a speech segment corresponding to frames that include only speech signals from among the frames based inclusive of the detected energy change.

A hand pattern switch device capable of easily and reliably detecting a hand pattern or a palm motion of an operator. From a picked-up image of a distal arm, an axis passing through the centroid of a hand is determined as a central axis passing through the center of arm. At least either first scanning lines perpendicular to the central axis or second scanning lines extending along the central axis are set between a fingertip and a palm center. While changing the scanning line to be examined from the fingertip side toward the palm center, a determination is made to determine how many number of scanning lines for each of which a finger width equal to or larger than a predetermined width is detected are present, thereby making a detection of whether or not the finger is extended from the palm.

Described is a computer-implemented system and method that dynamically detects which input device (e.g., pen or mouse) is currently in use, and based on the device, varies a program's user interaction model to better optimize the user's ability to interact with the program via that input device. A tablet input subsystem receives pen and touch data, and also obtains keyboard and mouse data. The subsystem analyzes the data and determines which input device is currently active. The active device is mapped to an interaction model, whereby different user interface appearances, behaviors and the like may be presented to the user to facilitate improved interaction. For example, a program may change the size of user interface elements to enable the user to more accurately scroll and make selections. Timing, tolerances and thresholds may change. Pen hovering can become active, and interaction events received at the same location can be handled differently.

A multi-functional remote controller (1) has a multi-functional button (20) consisting of an LCD panel (21), a touch panel (22), a transparent button sheet (23), and mechanical switches (22). The transparent button sheet (23) has a shape with concavities and convexities, and the shape enables a user to recognize the position of each button in the pattern using their sense of touch. When the user presses down the touch panel (22), a mechanical switch (24) is pressed down to give a click sensation.

The invention discloses a pedestrian re-identification method using attitude information to design a multi-loss function. The method can effectively solve the difficulties caused by frequent occlusionof pedestrians, large differences of video illumination and non-rigid pedestrian attitudes in surveillance video, and has a wide range of applications in security monitoring and other fields. The method is mainly divided into two phases, which are an offline phase and an online phase. The offline phase is responsible for training and learning a high-accuracy deep learning network model, and includes preprocessing, extraction of joint point information, extraction of local features and feature fusion with global features extracted by a backbone network framework, and finally training of the fused features using a quintuple loss function. In the online phase, the trained deep learning network model is used for feature extraction, so that pedestrian re-identification between a target to be analyzed and a stored target picture library is achieved by similarity calculation.

A display system and method created by adjusting the properties of one or more displays to obtain coarse control over display behavior, by using sensors to optimize display parameters and meet a quality target. The display may construct a display map by selectively driving the display and sensing the optical image created. The system reports on its status, and is able to predict when the system will no longer meet a quality target. The system and method may optimize a display system and keep it optimized over time. Operators of the display who require a minimum level of quality for the display system can be ensured that the display meets those requirements. And, they can be warned in advance as to when system maintenance can be necessary, when quality falls below targeted goals system and method provides for sending out methods of the quality of the system.

A device including a housing, a writing tip connected to the housing, a writing surface position indicator, a processor in the housing, a memory device in the housing connected to the processor, and a sensor in the housing and cooperative with the writing surface position indicator. The device may be used to record writings and drawings applied to a surface by a user, to transmit that data to a remote device, to download data from remote devices, and to otherwise communicate with remote devices.

A system and method of providing an autocomplete recommended word, which classify a recommended word list according to indexes of various languages, store the recommended word list for each index, extract a corresponding autocomplete recommended word according to a user query and a setting mode which is received from a user's web browser, provide the user with the corresponding autocomplete recommended word, and thereby may propose a suitable recommended word according to the user query.

A controlling system of a touch panel comprises of a drive signal generating circuit, a sensing module for sensing a waveform signal from the touch panel, and a signal processing module for enabling and disabling the drive signal generating circuit based on the signal outputted by the sensing module. Therefore, the mentioned controlling system can support the active and the passive touch modes. In addition, the present invention further provides a controlling method of the touch panel and a stylus pen applied to the mentioned controlling system.

A Bluetooth headset for a helmet having an inter-communication is provided. The Bluetooth headset includes a headset body configured to be mounted on a helmet, a microphone and a speaker that are configured to be installed on the headset body, a jog dial configured to be installed to perform a rotation operation and be subject to a push operation, a Bluetooth module configured to be accommodated in the headset body and to perform wireless communication while paired with an external Bluetooth device, and a control unit configured to be accommodated in the headset body to process a signal input from the microphone, a signal input according to the rotation operation and the push operation of the jog dial and a signal received from the Bluetooth module and output the processed signal to the speaker and the Bluetooth module.

A system is presented that provides real-time performance for iterative multi-user detectors, such as Turbo MUDs, which are used to separate simultaneous transmissions on the same frequency, by permitting the MUD to use a less computationally intense, fast-processing algorithm and to correct for errors caused by the fast processing. In order to reduce the errors, a voting system is coupled to the output of the multi-user detector within the iterative system. The voting system provides confidence values on a bit-by-bit basis for the estimates made by the multi-user detector, with the confidence values then being utilized as soft inputs to a bank of conventional single-user decoders.

An input section includes a palm rest and a hand detection sensor on the top surface of a housing. The input section also includes, for example, an analog resistive touch panel above the housing to the front. A palm rest is used as a base for placing a palm for manipulating the touch panel and has a flat surface slightly forward with reference to a horizontal level. The touch panel tilts below the surface of the palm rest. The touch panel is given a tilt angle so that the fingertip or the finger cushion naturally touches the surface (panel surface) of the touch panel when the palm is placed on the palm rest.

There is provided a data input device and a data input method. The data input device including: a sense unit provided by forming a predetermined sense region at a side of a case of an electronic apparatus terminal, for generating a first direction input signal by sensing a horizontal pressure in a predetermined radial direction of the finger put on a reference position within the sense region, and for generating a second direction input signal by sensing a vertical pressure leaned in a predetermined direction while the finger is put on the reference position; and a controller for determining a horizontal pressure direction or a vertical pressure position of the finger from the first direction input signal and the second direction input signal, and for extracting data assigned to a corresponding radial direction and a pressure position to input.

The invention relates to a human face value evaluation method, device and terminal device. The method comprises the following steps: conducting convolution processing on a human face image according to convolution layers of a convolution nerve network, so as to obtain local features of the human face image extracted from each convolution layer, wherein the human face image is the human face-containing area in an original image, and the convolution nerve network is subjected to a set number of task training; integrating the local features extracted from the convolution layers through a whole connection layer of a convolution nerve network, so as to link the local features into a one-dimensional vector with a set length; inputting the one-dimensional vector into a set number I of forecasting layers of the convolution nerve network respectively, so as to obtain a scoring value, relative to the human face, of the set number according to the forecasting layers. The method, device and terminal device have the advantages that a user can adjust a photographing process more specifically according to the scoring value, so as to improve the picture quality of a photograph to be shot or the skin smoothness of the user.

The invention provides a method for calculating gross vehicle weight. The method includes the steps that a gravity acceleration sensor, a controller and a vehicle original engine ECU which are installed on the vehicle are utilized, the controller figures out the current gross vehicle weight based on a vehicle driving kinetic equation by obtaining engine information, vehicle information and dip angle information and acceleration information and then outputs the current gross vehicle weight information outwards, wherein the dip angle information and the acceleration information are output by the gravity acceleration sensor. The vehicle information comprises the current gearbox gear speed ratio information, the final ratio information, the tier rolling radius information, the rolling damping coefficient information, the drag coefficient information, the conversion coefficient information of automobile rotary weight, transmission system mechanical efficiency information, clutch switching state information and the like. The invention further provides a fuel-saving control method. The method includes the steps that the current gross vehicle weight is obtained through calculation, the ratio of the current gross vehicle weight to vehicle full-load gross weight is converted into a loading state, and the ECU selects corresponding engine power output gears according to the received loading state information. The method for calculating the gross vehicle weight and the fuel-saving control method have the advantages that the number of used sensors is small, and calculation results are accurate.

The invention provides a multi-cycle installment decision making method and a multi-cycle installment decision making system. The multi-cycle installment decision making method comprises the steps of: acquiring user credit investigation information of a target user according to an installment request input by a user end or an external trigger, wherein the user credit investigation information comprises anti-fraud information and credit information of the user; applying for anti-fraud scanning based on the anti-fraud information and a called anti-fraud model, so as to obtain risk rating information; applying for credit scanning based on the credit information and a called credit model, so as to obtain credit rating information; and matching the risk rating information with a risk requirement and matching the credit rating information with a credit requirement in installment decision making, so as to output matched installment decisions. The multi-cycle installment decision making method and the multi-cycle installment decision making system can extract effective user information for data processing through classifying and integrating a large amount of user information, and feed back installment decision making schemes timely and accurately.

Contact time detection means (101) for detecting a time period for which a finger is in continual contact with a touch panel display (100), contact number detection means (102) for detecting the number of times a finger touched the touch panel display (100), and contact interval detection means (103) for detecting a time period between when the finger is detached form the touch panel display (100) and when the finger is brought into contact therewith afterwards are provided and based on these detection results, an input command is determined. Various commands can thereby be inputted without a visual check.

A variable capacitor is disclosed, including: a dielectric having a first surface and a second surface opposing the first surface; a first electrode disposed on the first surface of the dielectric; a second electrode disposed to face the second surface of the dielectric; and a pressing member configured to cause the second electrode and the dielectric to contact each other when a pressing force is applied thereto (e.g., when a pen-shaped position indicator including the variable capacitor is pressed against a tablet). In the variable capacitor, the second electrode includes at least one electrode piece having an abutting central portion, against which the pressing member transmits a pressing force applied thereto, and an extension portion that radially extends from the abutting central portion toward a circumferential portion of the dielectric. A contact area between the dielectric and the second electrode is changed in accordance with a pressing force applied to the pressing member, to thereby change a capacitance of the variable capacitor.

The present invention provides a suggestive and device of error correcting encode in the input course, the method includes following steps: receiving the encode string of user input; obtaining a corresponding candidate item by a conversion according to the received encode string; judging the candidate item whether exists candidate item obtained by easily-confused encode identification method or not; if yes, suggesting the error correcting encode information. In order to help the user input correct encode string, the invention creatively provides the suggestive information containing correct encoding string to the user under the confused condition, to help the user actively improve the input accuracy of the encode string in the use course, reducing the dependence of the fuzzy tone for user.

The invention discloses a deep learning-based advertisement click-through rate prediction method and device. The method includes the following steps that: a preset number of training advertisements as well as training click-through rates and training characteristics of each training advertisement are acquired; the training characteristics of each training advertisement are converted into training vectors, a deep learning model is trained by using the training vectors and the training click-through rates of each training advertisement, wherein the deep learning model is realized based on a nonlinear function; and a vector to be tested converted from characteristics to be tested of an advertisement to be tested is obtained, and the vector to be tested is adopted as the input of the deep learning model, and a predictive click-through rate corresponding to the advertisement to be tested is obtained. According to the deep learning model in the method of the invention, nonlinear relationships between the characteristics are fully considered, and thus, after the vector to be tested is inputted into the deep learning model, the deep learning model can efficiently and accurately output the predictive click-through rate corresponding to the vector to be tested based on the nonlinear function.

A method of displaying a display screen by using a display that is flexible is disclosed. The method includes detecting deformation of the display; determining whether an object is displayed within a set range from a folding line at which the deformation is detected; and modifying the displayed objected and displaying the modified displayed object when the object is displayed within the set range.

The invention discloses a commodity name recognition method based on a full-text attention mechanism. The method comprises the following steps: establishing a depth neural network model with a full-text attention mechanism; adding a full-text attention layer into the depth neural network model; in the full-text attention layer, the full-text information being represented by a method of embedding aword into a vector; the full-text attention layer being represented by a full-text information embedding vector. Then, for each word to be annotated, the similarity between them and the full-text attention-level word vector is calculated, so that the word can obtain the 'attention weight' of the full-text, that is, the word can focus on the context information of the document, and the additionalfeatures of each word can be calculated by using the sum of the attention-level weight, so as to identify the trade name. The method of the invention can effectively extract commodity names from irregular texts, solves the problem of inconsistent identification of the same commodity in context, and improves the accuracy of identification.