Abstract: An illumination system is configured to provide an illumination beam. The illumination system includes at least one light source, a composite unit and an actuator. The at least one light source is configured to provide at least one light beam. The composite unit includes a diffusion surface and an optical surface opposite to each other, wherein the composite unit is disposed on a transmission path of the at least one light beam and configured to enable the at least one light beam to pass through. The actuator is connected to the composite unit for driving the composite unit to move along at least one direction, and the at least one direction is perpendicular to a normal direction of a plane on which the composite unit is located. A projection apparatus including the illumination system is also provided.

Abstract: A method and apparatus include receiving a first sentence including a first set of words, and a second sentence including a second set of words. A first set of vectors corresponding to the first set of words of the first sentence, and a second set of vectors corresponding to the second set of words of the second sentence are generated using a word embedding model. A similarity matrix based on the first set of vectors and the second set of vectors is generated. An alignment score associated with the first set of vectors and the second set of vectors is determined using the similarity matrix. The alignment score is transmitted to permit information retrieval based on a similarity between the first sentence and the second sentence.

Abstract: A device and a method for improving a processing speed of a neural network and applications thereof in the neural network where the device includes a processor configured to perform: determining, according to a predetermined processing speed improvement target, a dimension reduction amount of each of one or more parameter matrixes in the neural network obtained through training; preprocessing each parameter matrix based on the dimension reduction amount of the parameter matrix; and retraining the neural network based on a result of the preprocessing to obtain one or more dimension reduced parameter matrixes so as to ensure performance of the neural network meets a predetermined requirement. According to the embodiments of the present disclosure, it is possible to significantly improve the processing speed of the neural network while ensuring the performance of the neural network meets the predetermined requirement.

Abstract: Methods, systems, and apparatus including computer programs encoded on a computer storage medium, for retrieving image search results using embedding neural network models. In one aspect, an image search query is received. A respective pair numeric embedding for each of a plurality of image-landing page pairs is determined. Each pair numeric embedding is a numeric representation in an embedding space. An image search query embedding neural network processes features of the image search query and generates a query numeric embedding. The query numeric embedding is a numeric representation of the image search query in the same embedding space. A subset of the image-landing page pairs having pair numeric embeddings that are closest to the query numeric embedding of the image search query in the embedding space are identified as first candidate image search results.

Abstract: A method, computer program, and computer system is provided for estimating three-dimensional hand poses in images. Data corresponding to two hand images is receive, and an optical flow value corresponding to a change in a hand gesture in the received hand image data is calculate. A heat map is generated based on the calculated optical flow, and a hand mesh map is estimated based on the generated heat map. A hand pose present within the hand images is determined based on the estimated hand mesh map.

Abstract: A semiconductor device includes a fin structure protruding from a first isolation insulating layer provided over a substrate, a gate dielectric layer disposed over a channel region of the fin structure, a gate electrode layer disposed over the gate dielectric layer, a base semiconductor epitaxial layer disposed over a source/drain region of the fin structure, and a cap semiconductor epitaxial layer disposed over the base semiconductor epitaxial layer. The cap semiconductor epitaxial layer has a different lattice constant than the base semiconductor epitaxial layer, and a surface roughness of the cap semiconductor epitaxial layer along a source-to-drain direction is greater than zero and smaller than a surface roughness of the base semiconductor epitaxial layer along the source-to-drain direction.

Abstract: This application discloses a method for predicting an attribute of a target object based on machine learning and a related device, which belong to the field of data prediction technologies. According to the method, a global feature of the target object is determined based on a rule feature representing historical and future change rules of a detection feature, and the global feature is refined to obtain at least one local feature of the target object, so that the refined local feature can better reflect the feature of the target object, and the attribute of the target object is further predicted based on the local feature, thereby improving the precision of the predicted attribute. When the attribute of the target object is a predicted diagnosis result, the precision of the predicted diagnosis result can be improved.

Abstract: A computer-implemented method, computer readable storage medium, and computer system is provided for estimating three-dimensional (3D) hand poses in images by receiving data corresponding to a hand image, generating a depth map corresponding to the received hand image data, and estimating a hand pose from the received hand image data and the generated depth map.

Abstract: The present disclosure is directed to source/drain (S/D) epitaxial structures with enlarged top surfaces. In some embodiments, the S/D epitaxial structures include a first crystalline epitaxial layer comprising facets; a non-crystalline epitaxial layer on the first crystalline layer; and a second crystalline epitaxial layer on the non-crystalline epitaxial layer, where the second crystalline epitaxial layer is substantially facet-free.

Abstract: An illumination system and a projection device, including a light emitting module, a light uniforming element, and a beam splitting/combining module, are provided. The light emitting module includes red and blue light elements. The light uniforming element has an optical axis and a light entrance surface. The beam splitting/combining module is disposed on a transmission path of at least one blue light beam, is located between the light emitting module and the light uniforming element, and includes at least one beam splitting element and at least one reflective element. At least one among the red and the blue light beams is reflected by the reflective element and the beam splitting element. The red and blue light beams are transmitted to the light uniforming element in a direction parallel to the optical axis, and a speckle distribution on the light entrance surface is symmetrical with respect to the optical axis.

Abstract: An illumination system and a projection device, including a light emitting module, a light uniforming element, and a beam splitting/combining module, are provided. The light emitting module includes red and blue light elements. The light uniforming element has an optical axis and a light entrance surface. The beam splitting/combining module is disposed on a transmission path of at least one blue light beam, is located between the light emitting module and the light uniforming element, and includes at least one beam splitting element and at least one reflective element. At least one among the red and the blue light beams is reflected by the reflective element and the beam splitting element. The red and blue light beams are transmitted to the light uniforming element in a direction parallel to the optical axis, and a speckle distribution on the light entrance surface is symmetrical with respect to the optical axis.

Abstract: An ultrasonic imaging system and an imaging method. The imaging method comprises: transmitting a divergent ultrasonic beam to a scanning object, and scanning the scanning object with the divergent ultrasonic beam (S11); a self-scanning object receiving an echo of the divergent ultrasonic beam, and obtaining divergent ultrasonic echo signals by means of beam synthesis (S12); obtaining blood flow velocity vector information of the scanning object according to the divergent ultrasonic echo signals (S13); and displaying the blood flow velocity vector information of the scanning object (S14). Using a divergent ultrasonic beam to perform blood flow imaging can ensure that there is a sufficiently large scanning area for covering a scanning object, thereby achieving ultrasonic blood flow imaging at a high frame rate.

Abstract: Systems and methods are disclosed for question generation to obtain more related medical information based on observed symptoms from a patient. In embodiments, possible diseases associated with the observed symptoms are generated by querying a knowledge graph. In embodiments, candidate symptoms associated with the possible diseases are also identified and are combined with the observed symptoms to obtain combined symptom sets. In embodiments, discriminative scores for the candidate symptom sets are determined and candidate symptoms with top discriminative scores are selected. In embodiments, these selected candidate symptoms may be checked for conflicts with observed symptoms and removed from further consideration if a conflict exists. In embodiments, one or more questions may be generated based on the remaining selected candidate systems to aid in collecting information about the patient. In embodiments, the process may be repeated with the updated observed symptoms.

Abstract: Presented are systems and methods that allow healthcare providers and governments to infer demand for healthcare resources to ensure effective and timely healthcare services to patients by reducing healthcare supply shortages, emergencies, and healthcare costs. In embodiments, this is accomplished by gathering data from a number of sources to generate labeled records from which entity features and relationships between entities are extracted, correlates, and/or combined with other external healthcare data. In embodiments, this information is used to train a model that predicts healthcare resource demands given a set of input conditions or factors.

Abstract: An unmanned aerial vehicle (UAV) includes a sensor configured to detect an indicator of a geo-fencing device; and a flight controller configured to generate one or more signals that cause the UAV to operate in accordance with a set of flight regulations that are generated based on the detected indicator of the geo-fencing device.

Abstract: A U-shaped unit and a liquid crystal element with U-shaped coplanar electrode units provided by the invention are capable of increasing a horizontal electric field intensity in a power supply state, so that when the invention is applied to be used as a liquid crystal driving element, a required horizontal electric field intensity can be achieved with a lower driving voltage to reduce a required driving power when the liquid crystal element is used as a display screen, thereby achieving an effect of power saving.

Abstract: Disclosed herein is a method of determining whether a subject has or is at risk of developing a cancer. The method comprises, obtaining a sample from the subject; determining the levels of at least two target polypeptides, which are selected from the group consisting of, ANXA2, HSPA5, KNG1 and MMP1; and assessing whether the subject has or is at risk of developing the cancer based on the levels of target polypeptides. The present method provides a potential means to diagnose and predict the occurrence of oral squamous cell carcinoma, and accordingly, the subject in need thereof could receive a suitable therapeutic regimen in time.

Abstract: A method and mobile terminal for processing data are disclosed. The method can include displaying a live streaming page, the live streaming page including a file sharing interface for guiding an anchor to select a file for sharing during live streaming; displaying a file list in the case that the file sharing interface is triggered; acquiring a target file selected from the file list; generating second live streaming data based on the target file, and displaying first live streaming data and the second live streaming data on the live streaming page; and synchronizing the first live streaming data and the second live streaming data to a second mobile terminal, wherein the first live streaming data is acquired by a camera of the first mobile terminal, and the second mobile terminal is an audience terminal.

Abstract: A high laser damage threshold reflective optically addressed liquid crystal spatial light modulator for shaping 1053 nm linearly polarized light beams, comprising a computer-controlled LCoS electrical addressable spatial light modulator, polarization beam splitter, and polarizer, Liquid crystal cell, analyzer, AC power supply, where the liquid crystal cell comprises a transparent conductive film antireflection film layer, a transparent conductive film base layer, a first transparent conductive layer, a liquid crystal alignment layer, a liquid crystal layer, an alignment element, a reflective film layer, a light guide layer, and a second transparent conductive layer. By changing the transparent conductive layer material of the light-transmitting part of the liquid crystal cell from ITO to gallium nitride material, the damage threshold of the high-energy laser is improved, which facilitates application of beam shaping in high-power laser devices.

Abstract: A method and apparatus for generating augmented training data for hand pose estimation include receiving source data that is associated with a first lighting condition. Target data that is associated with a second lighting condition is received. A lighting condition translation between the first lighting condition and the second lighting condition is determined. Lighting translated data is generated based on the lighting condition translation and the source data. Augmented training data for hand pose estimation is generated based on the target data and the lighting translated data.