635 results about "Dynamic feature" patented technology

An apparatus for providing voice conversion using temporal dynamic features includes a feature extractor and a transformation element. The feature extractor may be configured to extract dynamic feature vectors from source speech. The transformation element may be in communication with the feature extractor and configured to apply a first conversion function to a signal including the extracted dynamic feature vectors to produce converted dynamic feature vectors. The first conversion function may have been trained using at least dynamic feature data associated with training source speech and training target speech. The transformation element may be further configured to produce converted speech based on an output of applying the first conversion function.

An apparatus for providing voice conversion using temporal dynamic features includes a feature extractor and a transformation element. The feature extractor may be configured to extract dynamic feature vectors from source speech. The transformation element may be in communication with the feature extractor and configured to apply a first conversion function to a signal including the extracted dynamic feature vectors to produce converted dynamic feature vectors. The first conversion function may have been trained using at least dynamic feature data associated with training source speech and training target speech. The transformation element may be further configured to produce converted speech based on an output of applying the first conversion function.

A system and a method for phrase-based translation are disclosed. The method includes receiving source language text to be translated into target language text. One or more dynamic bi-phrases are generated, based on the source text and the application of one or more rules, which may be based on user descriptions. A dynamic feature value is associated with each of the dynamic bi-phrases. For a sentence of the source text, static bi-phrases are retrieved from a bi-phrase table, each of the static bi-phrases being associated with one or more values of static features. Any of the dynamic bi-phrases which each cover at least one word of the source text are also retrieved, which together form a set of active bi-phrases. Translation hypotheses are generated using active bi-phrases from the set and scored with a translation scoring model which takes into account the static and dynamic feature values of the bi-phrases used in the respective hypothesis. A translation, based on the hypothesis scores, is then output.

An electrophysiology (EP) system includes an interface for operator-interaction with the results of code executing therein. A template model can have channels positioned or repositioned thereupon by the user to define a set-up useful in rapid catheter positioning. Mapping operations can be performed without the requirement for precise catheter location determinations. A map module coordinates EP data associated with each selected channel and its associated position on the template model to provide this result, and can update the resulting map in the event that the channel or location is changed. Messaging and other dynamic features enable synchronized presentation of a myriad of EP data. Additional systems and methods are disclosed herein.

A method is described for calculating a backup connection for restoring a nominal connection in a telecommunication network. The method performs the calculation of the backup connection taking into account updated values of parameters indicating at least one dynamic feature of a group of segments of the network. The dynamic parameters can be a measure of the performance of the segment, the degree of usage of the segment, the bandwidth overbooking of the segment or the bandwidth fragmentation or a combination of these measures.

A document annotation method includes modeling data elements of an input document and dependencies between the data elements as a dependency network. Static features of at least some of the data elements are defined, each expressing a relationship between a characteristic of the data element and its label. Dynamic features are defined which define links between an element and labels of the element and of a second element. Parameters of a collective probabilistic model for the document are learned, each expressing a conditional probability that a first data element should be labeled with information derived from a label of a neighbor data element linked to the first data element by a dynamic feature. The learning includes decomposing a globally trained model into a set of local learning models. The local learning models each employ static features to generate estimations of the neighbor element labels for at least one of the data elements.

A system and method for determining the temporal resolution of a tomographic imaging device uses an apparatus to drive one or more dynamic phantoms composed of multiple materials. The apparatus is placed at or near the isocenter of the imaging device and the one or more phantoms are moved to produce a plurality of dynamic features, each having a specified frequency. The dynamic features are imaged with the device and the acquired image data corresponding to the dynamic features is analyzed to determine a temporal modulation transfer value at each of the known specified frequencies. The temporal resolution of the imaging device is determined using these temporal modulation transfer values.

The invention discloses an Android malicious application detection method and system based on multi-feature fusion. The method comprises the following steps that: carrying out decompilation on an Android application sample to obtain a decompilation file; extracting static features from the decompilation file; operating the Android application sample in an Android simulator to extract dynamic features; carrying out feature mapping on the static features and the dynamic features by the text Hash mapping part of a locality sensitive Hash algorithm, mapping to a low-dimensional feature space to obtain a fused feature vector; and on the basis of the fused feature vector, utilizing a machine learning classification algorithm to train to obtain a classifier, and utilizing the classifier to carry out classification detection. By use of the method, the high-dimensional feature analysis problem of a malicious code rare sample family can be solved, and detection accuracy is improved.

A method to determine the propensity of an image-sequence to induce motion sickness in a viewer includes using a processor to analyze the image-sequence information to extract salient static and dynamic visual features in the image sequence information, evaluating distribution of the salient static and dynamic features in the saliency map to estimate a probability of the image sequence causing the viewer to make eye and head movements and, using the estimated probability to determine the propensity that the image-sequence would induce motion sickness in a user as a consequence of the eye and head movements.

Described are a method and a system for localization of a vehicle. The method includes the acquisition of SPR images of a subsurface region along a vehicle track. Acquired SPR images are compared to SPR images previously acquired for a subsurface region that at least partially overlaps the subsurface region along the vehicle track. In some embodiments, the comparison includes an image correlation procedure. Location data for the vehicle are determined based in part on location data for the SPR images previously acquired for the second subsurface region. Location data can be used to guide the vehicle along a desired path. The relatively static nature of features in the subsurface region provides the method with advantages over other sensor-based navigation systems that may be adversely affected by weather conditions, dynamic features and time-varying illumination. The method can be used in a variety of applications, including self-driving automobiles and autonomous platforms.

There is provided a singing synthesis parameter data estimation system that automatically estimates singing synthesis parameter data for automatically synthesizing a human-like singing voice from an audio signal of input singing voice. A pitch parameter estimating section 9 estimates a pitch parameter, by which the pitch feature of an audio signal of synthesized singing voice is got closer to the pitch feature of the audio signal of input singing voice based on at least both of the pitch feature and lyric data with specified syllable bondaries of the audio signal of input singing voice. A dynamics parameter estimating section 11 converts the dynamics feature of the audio signal of input singing voice to a relative value with respect to the dynamics feature of the audio signal of synthesized singing voice, and estimates a dynamics parameter, by which the dynamics feature of the audio signal of synthesized singing voice is got close to the dynamics feature of the audio signal of input singing voice that has been converted to the relative value.

The present invention is a method and system for automatically recognizing which products a shopper intends to find or purchase based on the shopper's trajectory in a retail aisle. First, the system detects and tracks the person to generate the trajectory of the shopper. Then some of the dynamic features are extracted from the shopper trajectory. The shopper trajectory features of a given trajectory are typically the positions, the motion orientations, and speeds at each point of the trajectory. A shopper behavior model is designed based on some of the primitive actions of shoppers. The last step of the method is to analyze a given shopper trajectory to estimate the shopper's intention. The step either utilizes decision rules based on the extracted shopper trajectory features, or utilizes a trained Hidden Markov Model, to estimate the progression of the primitive actions from the trajectory. The decoded progression of the shopper behavior states is then interpreted to finally determine the shopper's intention.

The invention discloses a detecting system for an Android malicious code based on deep learning. The detecting system comprises a feature extracting module, a deep learning module and a report generating module, wherein the feature extracting module is used for taking APK (Android Package) procedure as input, combining the static extraction with the dynamic extraction and outputting 0 and 1 for forming an APK procedure feature vector; the deep learning module is used for taking a MLP (Modular Longitudinal Platform) model as a learning model, on the one hand, training and learning a sample set formed by the feature vector and a supervised value, thereby acquiring a mature learning model; on the other hand, the deep learning module is used for taking the feature vector as input, using the mature learning model for outputting a result probability and taking the result probability as the security level of the APK procedure; the report generating module is used for explaining and analyzing for lastly generating an assessment report according to the feature vector and security level of the APK procedure. According to the detecting system provided by the invention, the deep learning module is combined with the detection for the malicious code while the static feature is combined with the dynamic feature, so that the discriminating capability and detecting accuracy of the system for the unknown malicious code are promoted.

3-dimensional cardiac reconstruction is carried out by catheterizing a heart using a probe with a mapping electrode, and acquiring electrical data from respective locations in regions of interest in the heart, representing the locations of the electrical data as a point cloud, reconstructing a model of the heart from the point cloud, applying a set of filters to the model to produce a filtered volume, segmenting the filtered volume to define components of the heart, and reporting the segmented filtered volume.

The present invention is directed to systems and methods for controlling 2-D to 3-D image conversion. The system and method includes receiving an image and masking the objects in the image using segmentation layers Each segmentation layer can have weighted values for static and dynamic features. Iterations are used to form the final image which, if desired, can be formed using less than all of the segmentation layers. A final iteration can be run with the weighted values equal for static and dynamic features.

Described are a method and a system for localization of a vehicle. The method includes the acquisition of SPR images of a subsurface region along a vehicle track. Acquired SPR images are compared to SPR images previously acquired for a subsurface region that at least partially overlaps the subsurface region along the vehicle track. In some embodiments, the comparison includes an image correlation procedure. Location data for the vehicle are determined based in part on location data for the SPR images previously acquired for the second subsurface region. Location data can be used to guide the vehicle along a desired path. The relatively static nature of features in the subsurface region provides the method with advantages over other sensor-based navigation systems that may be adversely affected by weather conditions, dynamic features and time-varying illumination. The method can be used in a variety of applications, including self-driving automobiles and autonomous platforms.

A dynamic and differential CMOS logic style is disclosed in which a gate uses a fixed amount of energy per evaluation event. The gate switches its output at every event and loads a constant capacitance. The logic style is a Dynamic and Differential Logic (DDL) style. The DDL style logic typically has one charging event per clock cycle and the charging event does not depend on the input signals. The differential feature masks the in-put value because a precharged output nodes is discharged during the evaluation phase. The dynamic feature breaks the input sequence: the discharged node is charged during the subsequent precharge phase.

The invention provides an implement method and a terminal for dynamic function menus. The method comprises the steps: counting usage frequency of each function menu by users, and obtaining counted information of the function menus; and adjusting a list or sequence of a layout of the function menus in a dynamic mode according to the counted information of the function menus and based on a preset arrangement form or a layout form. According to the implement method and the terminal for the dynamic function menus, the function menus can be properly adjusted on arrangement or layout according to using habits or form requirements of the users, so that usage experience on the menus is improved, and convenience is brought for operation by the users.

A system and a method for phrase-based translation are disclosed. The method includes receiving source language text to be translated into target language text. One or more dynamic bi-phrases are generated, based on the source text and the application of one or more rules, which may be based on user descriptions. A dynamic feature value is associated with each of the dynamic bi-phrases. For a sentence of the source text, static bi-phrases are retrieved from a bi-phrase table, each of the static bi-phrases being associated with one or more values of static features. Any of the dynamic bi-phrases which each cover at least one word of the source text are also retrieved, which together form a set of active bi-phrases. Translation hypotheses are generated using active bi-phrases from the set and scored with a translation scoring model which takes into account the static and dynamic feature values of the bi-phrases used in the respective hypothesis. A translation, based on the hypothesis scores, is then output.

The invention relates to parallel robot industry, comprising dynamic and static platforms and three branches connected to them. I features in the dynamic platform through three branches and connected to the static platform, with one branch having a movement sub, the second branch having three movement subs, the first and the second branches forming into horizontal guide pole structure or plane hinge structure, with the third branch having space for six freedom branch chains. It enjoys high rigidity, simple structure, two rotary complete decoupling, easy for control, low in movement quality, good in dynamic features.

A computer implemented method and a mobile stamp creation and communication system (MSCCS) dynamically creates, renders, and transforms a digital stamp for a stamped communication and media interactions. The MSCCS creates digital stamps by integrating media content, data elements, and dynamic functional elements received from multiple sources. The MSCCS renders the digital stamps to a user device for user selection based on dynamic rendering criteria, generates a stamped communication with the selected digital stamps, and marks the selected digital stamps with a visual usage marker that indicates use of the selected digital stamps in the stamped communication. The MSCCS transforms the marked digital stamps into dynamic content on the user device using data elements and functional elements imbedded in the marked digital stamps and inherent elements of the user device for performing actions on the marked digital stamps and invoking media interactions for creation and rendering of enhanced digital stamps.

3-dimensional cardiac reconstruction is carried out by catheterizing a heart using a probe with a mapping electrode, and acquiring electrical data from respective locations in regions of interest in the heart, representing the locations of the electrical data as a point cloud, reconstructing a model of the heart from the point cloud, applying a set of filters to the model to produce a filtered volume, segmenting the filtered volume to define components of the heart, and reporting the segmented filtered volume.

The present invention generally is directed to a system, method and article of manufacture for accessing data independent of the particular manner in which the data is physically represented. In one embodiment, a data repository abstraction layer provides a logical view of the underlying data repository that is independent of the particular manner of data representation. A query abstraction layer is also provided and is based on the data repository abstraction layer. A runtime component performs translation of an abstract query into a form that can be used against a particular physical data representation.

The invention discloses a voice converting method based on deep learning, and belongs to the technical field of voice signal processing. According to the invention, the method includes the following steps: configuring a voice encoding and decoding device AHOcoder as a feature extraction terminal and a voice synthesis terminal, training voice features by using the deep leaning method to separatelyobtain deep features of a source speaker and a target speaker, also obtaining the capability of decoding the deep features to original features, mapping the source speaker and the target speaker by using a BP neural network, thus realizing voice conversion. According to the invention, the method stitches the original features of voice, the combined feature parameters obtained from stitching are deemed to include the dynamic features of the voice features of the speaker, the training of the deep neural network is accelerated by pre-training the deep autoencoders, and by converting the deep features, the method herein obtains quality converted voice even when less voice materials are trained. The method also supports offline learning, and saves computing resources and memory of terminal devices.

The invention discloses an anomaly detection method based on network flow analysis. A complete network flow initial feature set is provided by deeply analyzing an IP (internet protocol) data packet, and the performance of an anomaly detection system can be fundamentally improved. A feature subset for anomaly detection is dynamically selected according to different types of network anomalies, finally, the class of an unknown sample is predicted according to the feature subset by a Bayes classifier, and the anomalies are prompted if the unknown sample is abnormal according to a prediction result. A data preprocessing module is used for processing preliminary data; a feature selecting module is used for selecting the proper feature subset for anomaly detection according to the types of the anomalies; and an anomaly detection module is used for prompting the anomalies after the anomalies are found. By the aid of dynamic feature selection algorithm, the optimal feature subset for detecting the anomalies can be dynamically selected according to the different types of anomalies, flow feature dimensions for detecting the anomalies can be reduced, and anomaly detection accuracy is improved.

The invention relates to a virtual network function dynamic migration method based on deep belief network resource demand forecasting, and belongs to the field of mobile communication. The method comprises the following steps: (S1) in view of the dynamic features of SFC business resource demand in a slicing network, establishing a system overhead model of comprehensive migration overhead and bandwidth overhead; (S2) in order to realize spontaneous VNF migration, monitoring the resource utilization condition of virtual network function or link in real time, and discovering the deployed bottom nodes or resource hot spots in the link in time by using an online learning based adaptive DBN forecasting method; (S3) designing a topology awareness based dynamic migration method according to the forecasting result, so as to reduce system overhead; (S4) proposing a tabu search based optimization method to further optimize the migration strategy. The forecasting method provided by the invention not only increase the convergence rate of a training network, but also realizes a perfect forecasting effect; by combining the forecasting method with a migration method, the system overhead and the violation frequency of the service level agreement are effectively reduced, and the performance of network service is improved.

A dynamic and differential CMOS logic style is disclosed in which a gate uses a fixed amount of energy per evaluation event. The gate switches its output at every event and loads a constant capacitance. The logic style is a Dynamic and Differential Logic (DDL) style. The DDL style logic typically has one charging event per clock cycle and the charging event does not depend on the input signals. The differential feature masks the in-put value because a precharged output nodes is discharged during the evaluation phase. The dynamic feature breaks the input sequence: the discharged node is charged during the subsequent precharge phase.