Abstract: The present disclosure discloses a text classification system based on feature selection and a method thereof in the technical field of natural language processing and short text classification, comprising: acquiring a text classification data set; dividing the text classification data set into a training text set and a test text set, and then pre-processing the training text set and the test text set; extracting feature entries from the pre-processed training text set through improved chi-square statistics to form feature subsets; using TF-IWF algorithm to give the weight to the extracted feature entries; based on the weighted feature entries, establishing a short text classification model based on a support vector machine; and classifying the pre-processed test text set by the short text classification model. The present disclosure solves the problem that the short text content is sparse to some extent, thereby improving the performance of short text classification.

Abstract: A configuration for reporting OOD samples for neural network optimization. The apparatus receives, from a base station, a configuration to report an OOD dataset for a machine learning model. The apparatus detects an occurrence of one or more OOD events. The apparatus reports the OOD dataset comprising the one or more OOD events based on the configuration to report OOD dataset. The apparatus receives, from the base station, an update to the machine learning model. The OOD dataset may comprise raw data related to the one or more OOD events, or may comprise extracted latent data corresponding to features of raw data related to the one or more OOD events.

Abstract: Systems and methods enable machine learning (ML) feature prioritization to reduce image noise in a virtual environment. In embodiments, a method includes providing a device access to a virtual environment via a graphical user interface (GUI), the environment including images of objects and a navigation tool enabling a user to navigate the environment and interact with the images; monitoring interaction data of the user; calculating priority values for predefined areas of a first object in the environment, using an ML model trained with historic user interaction data and object data; processing image data of one or more of the predefined areas of the first object using image processing to generate new data based on display specifications of the client device and the priority values; and pre-loading the new data in a buffer, such that the new data is available prior to display of the new data to the user.

Abstract: A device includes: a complementary transistor including: a first transistor having a first source/drain region and a second source/drain region; and a second transistor stacked on the first transistor, and having a third source/drain region and a fourth source/drain region, the third source/drain region overlapping the first source/drain region, the fourth source/drain region overlapping the second source/drain region. The device further includes: a first source/drain contact electrically coupled to the third source/drain region; a second source/drain contact electrically coupled to the second source/drain region; a gate isolation structure adjacent the first and second transistors; and an interconnect structure electrically coupled to the first source/drain contact and the second source/drain contact.

Abstract: An LED light source device includes a first substrate, which is light transmittable and includes a first joining surface; at least two terminal pins, each of which includes an internal connection portion connected to a first joining surface of the first substrate and an external connection portion exposed outside the first joining surface; a control chip and an LED chip mounted to the first joining surface of the first substrate; a conductive wire mounted to the first joining surface of the first substrate and connected to the control chip, the LED chip, and the terminal pins; a second substrate, which is light transmittable and includes a second joining surface laminated on the first joining surface of the first substrate to have the conductive wire, the control chip, the LED chip, and the internal connection portions of the terminal pins enclosed between the first joining surface and the second joining surface.

Abstract: A one-hand bit releasing hand tool includes a hand tool main body having a first drive section, which has a first drive hole and a receiving hole with a stepped portion arranged therebetween; and a positioning and releasing assembly, which includes a positioning member disposed in the receiving hole and supported on the stepped portion to apply an action force on and thus position a bit placed into the first drive hole and a releasing member disposed in the receiving hole and reciprocally movable between a positioning position and a releasing position. When the bit is placed into the first drive hole, the releasing member moves from the releasing position to the positioning position; and when the releasing member is moved from the positioning position to the releasing position, the releasing member resists the action force of the positioning member to release the bit.

Abstract: A method, system, and computer program product controlling flash behavior of a camera during capture of image data. The method includes identifying a target object from among a plurality of objects in a current scene. The method further includes determining whether a distance to the target object is within an effective flash range associated with at least one flash module of an image capturing device. The method further includes measuring, via at least one camera, an ambient light level within the current scene and comparing the measured ambient light level to a predetermined flash light threshold. The method further includes, in response to determining the target object is within the effective flash range and the measured ambient level is less than the predetermined flash light threshold, triggering at least one flash module of the image capturing device to illuminate the current scene during capture of image data by the at least one camera.

Abstract: A method, a system, and a computer program product for dynamically adjusting sampling of a depth map based on detected motion in a current scene. The method includes capturing real-time scan data at a first resolution by a first camera and a second camera of an image capturing device. The method further includes synchronizing a first plurality of frames of the real-time scan data to create a plurality of synchronized frames at a first frame rate. The method further includes analyzing the synchronized frames to determine whether motion exists. The method further includes, in response to determining motion exists: determining, based on the plurality of synchronized frames, a rate of motion within the current scene; and dynamically calculating a target resolution and a target frame rate for a real-time depth map. The method further includes generating a real-time depth map at the target resolution and target frame rate.

Abstract: A method, system, and computer program product for generating a high dynamic range (HDR) image from a plurality of frames simultaneously captured by cameras of a portable device. The method includes detecting, by a main camera of the portable device, a current dynamic range level of a current scene. In response to detecting the current dynamic range level, a main camera exposure compensation value is applied to the main camera and an auxiliary camera exposure compensation value, which is different from the main camera exposure compensation value, is applied to at least one auxiliary camera. The method further includes simultaneously capturing a first frame by the main camera and at least one second frame by at least one auxiliary camera. The method further includes combining the first frame and the at least one second frame to create the HDR image.

Abstract: A method, a system, and a computer program product for dynamically adjusting sampling of a depth map based on detected motion in a current scene. The method includes capturing real-time scan data at a first resolution by a first camera and a second camera of an image capturing device. The method further includes synchronizing a first plurality of frames of the real-time scan data to create a plurality of synchronized frames at a first frame rate. The method further includes analyzing the synchronized frames to determine whether motion exists. The method further includes, in response to determining motion exists: determining, based on the plurality of synchronized frames, a rate of motion within the current scene; and dynamically calculating a target resolution and a target frame rate for a real-time depth map. The method further includes generating a real-time depth map at the target resolution and target frame rate.

Abstract: A method, system, and computer program product controlling flash behavior of a camera during capture of image data. The method includes identifying a target object from among a plurality of objects in a current scene. The method further includes determining whether a distance to the target object is within an effective flash range associated with at least one flash module of an image capturing device. The method further includes measuring, via at least one camera, an ambient light level within the current scene and comparing the measured ambient light level to a predetermined flash light threshold. The method further includes, in response to determining the target object is within the effective flash range and the measured ambient level is less than the predetermined flash light threshold, triggering at least one flash module of the image capturing device to illuminate the current scene during capture of image data by the at least one camera.

Abstract: An electronic device, computer program product and method provide automatic exposure control (AEC) convergence in a dual camera system. A main camera is triggered to perform AEC convergence procedure at a first frame rate. A display presents at least one of first images from the main camera. An auxiliary camera is triggered to perform AEC convergence procedure. A determination is made whether the AEC of the auxiliary camera has converged. In response convergence by the AEC, an exposure setting of the main camera is synchronized to the exposure setting of the auxiliary camera. The display presents a next image captured by the main camera operating with the synchronized exposure settings.

Abstract: An electronic device, computer program product and method provide automatic exposure control (AEC) convergence in a dual camera system. A main camera is triggered to perform AEC convergence procedure at a first frame rate. A display presents at least one of first images from the main camera. An auxiliary camera is triggered to perform AEC convergence procedure. A determination is made whether the AEC of the auxiliary camera has converged. In response convergence by the AEC, an exposure setting of the main camera is synchronized to the exposure setting of the auxiliary camera. The display presents a next image captured by the main camera operating with the synchronized exposure settings.

Abstract: A method and apparatus provide automatic stage scene detection. A first image of a scene can be captured using a luminance for automatic exposure control. A brightness contrast between a center region of the first image and areas surrounding the center region of the first image can be calculated. A brightness value can be ascertained for the entire first image. Whether the luminance of the first image should be adjusted can be determined based on the brightness contrast being greater than a contrast threshold and the brightness value for the entire first image being less than a brightness value threshold. The luminance for the automatic exposure control can be adjusted for correct exposure of the center region of the first image in response to determining the luminance of the first image should be adjusted. A second image of the scene can be captured based on automatic exposure control using the adjusted luminance.

Abstract: For backlit face detection, an apparatus includes a main camera and an auxiliary camera that capture image data of photographic subject matter. The main camera and auxiliary camera are located on the same surface of the apparatus. The apparatus further includes a processor that increases an exposure level of the auxiliary camera to expose a dark region in response to the photographic subject matter being a backlit scene. Here settings the main camera are unaffected by increasing the exposure level of the auxiliary camera. The processor also searches for a face within the dark region using auxiliary image data captured with the increased exposure level. Additionally, the processor adjusts an exposure setting of the main camera based on face information received from the auxiliary camera in response to a face being present within the dark region.

Abstract: A method, system, and computer program product for generating a high dynamic range (HDR) image from a plurality of frames simultaneously captured by cameras of a portable device. The method includes detecting, by a main camera of the portable device, a current dynamic range level of a current scene. In response to detecting the current dynamic range level, a main camera exposure compensation value is applied to the main camera and an auxiliary camera exposure compensation value, which is different from the main camera exposure compensation value, is applied to at least one auxiliary camera. The method further includes simultaneously capturing a first frame by the main camera and at least one second frame by at least one auxiliary camera. The method further includes combining the first frame and the at least one second frame to create the HDR image.

Abstract: A method and apparatus provide automatic stage scene detection. A first image of a scene can be captured using a luminance for automatic exposure control. A brightness contrast between a center region of the first image and areas surrounding the center region of the first image can be calculated. A brightness value can be ascertained for the entire first image. Whether the luminance of the first image should be adjusted can be determined based on the brightness contrast being greater than a contrast threshold and the brightness value for the entire first image being less than a brightness value threshold. The luminance for the automatic exposure control can be adjusted for correct exposure of the center region of the first image in response to determining the luminance of the first image should be adjusted. A second image of the scene can be captured based on automatic exposure control using the adjusted luminance.

Abstract: A method and apparatus provide automatic stage scene detection. A first image of a scene can be captured using a frame luma for automatic exposure control. A brightness contrast can be calculated between a center region of the first image and areas surrounding the center region of the first image. A brightness value can be ascertained for the entire first image. The first image can be determined to be a stage scene image based on the brightness contrast being greater than a stage scene contrast threshold and the brightness value for the entire first image being less than a stage scene brightness value threshold. The frame luma for the automatic exposure control can be adjusted for correct exposure of the center region of the first image in response to determining the first image is a stage scene.

Abstract: A method and apparatus provide automatic stage scene detection. A first image of a scene can be captured using a frame luma for automatic exposure control. A brightness contrast can be calculated between a center region of the first image and areas surrounding the center region of the first image. A brightness value can be ascertained for the entire first image. The first image can be determined to be a stage scene image based on the brightness contrast being greater than a stage scene contrast threshold and the brightness value for the entire first image being less than a stage scene brightness value threshold. The frame luma for the automatic exposure control can be adjusted for correct exposure of the center region of the first image in response to determining the first image is a stage scene.

Abstract: Financial advisory methods and systems for creating a steady lifetime income stream within an investment plan is provided. According to one embodiment, a feasible, personalized pattern of periodic cash payouts from an investment plan is structured to achieve level annual income during a payout period when combined with other income sources. A mapping is maintained of fixed income investments from a limited universe of financial products available within the investment plan to corresponding constant maturity Treasury bond (CMT) weights. Based on the mapping, a payout program is designed that is expected to generate a stream of income sufficient to fund the cash payouts by determining an appropriate allocation of assets within the investment plan among the fixed income investments. A payout portfolio is formed within the investment plan including holdings in the fixed income investments in accordance with the determined appropriate allocation. The cash payouts are paid to the investor.