Abstract: An exemplary embodiment relates to the field of Automatic Face Recognition (AFR) systems. More specifically one exemplary embodiment relates at least to a method and a system capable of recognizing the face of a person using a device equipped with a camera of any kind and an associated computer, such as an embedded computer. The system is alternatively suitable to be implemented as an embedded system with minimal processing hardware capabilities, consuming very low power.

Abstract: Systems and methods for adapting physical activities and exercises based on facial analysis by image processing are disclosed. A method includes: identifying, by a computer device, a user; receiving, by the computer device, video data of a face region of the user while the user is engaged in exercise or physical activity; analyzing, by the computer device, the video data to determine a detected state of the user, wherein the analyzing the video data includes performing facial analysis using the video data; and providing, by the computer device, feedback to the user based on the analyzing the video data.

Abstract: The present invention discloses an expression recognition apparatus and methods using a head mounted display thereof. A head mounted display apparatus for performing expression recognition according to an embodiment of the present invention comprises a sensing unit including at least one of expression detection sensing units installed inside of said apparatus for sensing expression information around eyes by at least one of a contact or a non-contact manner; an image acquiring unit installed outside of said apparatus for collecting expression information around a mouth; and an acquisition unit for information of expressions for collecting expression information around the eyes and expression information around the mouth.

Abstract: A method and apparatus for deep learning based fine-grained body part recognition in medical imaging data is disclosed. A paired convolutional neural network (P-CNN) for slice ordering is trained based on unlabeled training medical image volumes. A convolutional neural network (CNN) for fine-grained body part recognition is trained by fine-tuning learned weights of the trained P-CNN for slice ordering. The CNN for fine-grained body part recognition is trained to calculate, for an input transversal slice of a medical imaging volume, a normalized height score indicating a normalized height of the input transversal slice in the human body.

Abstract: Systems and methods are disclosed herein for annotating video tracks obtained from video data streams. Video tracks are treated as positive if they contain at least one region of interest containing a particular person, and negative if the video track does not contain a region of interest containing the particular person. Visual similarity models are trained using the positive bags.

Abstract: A method for processing an image with a processor, including: identifying, within an image, a candidate arrow including a long path adjacent to a plurality of short paths; determining a direction of the candidate arrow based on the long path; estimating a tip pixel within the plurality of short paths based on the direction of the candidate arrow; determining, for the candidate arrow, a plurality of arrowhead edges in the plurality of short paths based on the tip pixel; and generating, for the candidate arrow, a plurality of refined arrowhead edges based on the plurality of arrowhead edges, the direction, and the tip pixel.

Abstract: A method, apparatus and computer program product are provided for object identification and location correlation based on received images. A method is provided including receiving identity data associated with an object derived from one or more of a plurality of images determining an identity of an object in one or more of the plurality of images, receiving location information associated with a camera which captured the plurality of images, and correlating the identified object with a location for each of the respective images of the plurality of images.

Abstract: A camera system comprises an image capturing device, object detection module, object tracking module, and match classifier. The object detection module receives image data and detects objects appearing in one or more of the images. The object tracking module temporally associates instances of detected objects, each of which has a signature representing features of the detected object. The match classifier matches object instances by analyzing data derived from the signatures. The match classifier determines whether the signatures match.

Abstract: A simple and quick verification of the authenticity of a blister of elements, such as pills in the pharmaceutical industry, is provided. The pills packaged in a blister can be authenticated using mobile imaging devices such as a mobile smartphone. Generally speaking, an image of the pills and their differentiating physical characteristics in the blister is acquired and compared to at least one code which is marked on the blister itself. This code may also be encrypted. A dedicated software application on the mobile smartphone can be used to decrypt the code and compare the code to the image of the pills in the blister, to thereby evaluate authenticity of the blister and its contents.

Abstract: Methods and systems are disclosed including a computerized system, comprising: a computer system having an input unit, a display unit, one or more processors, and one or more non-transitory computer readable medium, the one or more processors executing software to cause the one or more processors to: display on the display unit one or more images, from an image database, depicting a structure; receive a validation from the input unit indicating a validation of a location of the structure depicted in the one or more images; generate unmanned aircraft information including flight path information configured to direct an unmanned aircraft to fly a flight path above the structure and capture sensor data from a camera on the unmanned aircraft while the unmanned aircraft is flying the flight path; receive the sensor data from the unmanned aircraft; and generate a structure report based at least in part on the sensor data.

Abstract: Access to a set of media content items is acquirable. Identified processors can perform, in parallel, object detection for the set. In some cases, information about a current system state, a user, and/or object popularity metrics is acquirable for selecting a subset of object models. Object recognition is performable, based on the subset, for the set of media content items. In some instances, a camera view can be provided. Object recognition is performable for representations of the view. An object depicted in the representations is identifiable. An interface portion is presentable to provide a label for the object. In some cases, object recognition is performable for the set of media content items to identify an object depicted in a content item. A label is associable with the content item. A search through the set of media content items can identify, based on the label, a subset that depicts the object.

Abstract: The present disclosure relates to navigation field and discloses a blindman navigation method and a blindman navigation apparatus. The blindman navigation method includes: obtaining image information of surroundings where a user is located; determining a distance between the user and a traffic signal light according to the image information; judging a distance between the user and a crossing according to the distance between the user and the traffic signal light, and determining the user is in a state where he/she has arrived at the crossing when the distance between the user and the crossing is less than a distance threshold that is preset.

Abstract: [Object] To link devices by recognizing the appearance of a device, such as an unknown application or a dynamically changing application. [Solution] An information processing apparatus according to the present disclosure includes: an appearance information acquisition unit configured to acquire appearance information indicating a feature of appearance of an own device; and a sending unit configured to send the appearance information to communicate with another device that has imaged the appearance of the own device. According to such a configuration, it is possible to link devices by recognizing the appearance of a device, such as an unknown application or a dynamically changing application.

Abstract: The present disclosure provides a system for an object recognition system for real time recognition of one or more objects captured in an image of one or more images. The object recognition system includes a first step of receiving the one or more images of the one or more objects. In addition, the object recognition system includes another step of analyzing each image of the one or more images. Further, the object recognition system includes yet another step of creating one or more models. Furthermore, the object recognition system includes yet another step of segmenting. The object recognition system includes yet another step of matching the one or more segmented objects with the one or more models. The object recognition system includes yet another step of recognizing the one or more objects. The object recognition system includes displays one or more information. The object recognition system calculates a probability score.

Abstract: Systems and methods are provided for providing an instruction set to a user for assembling, repairing, or maintaining a manufacturable entity. The method includes utilizing an input device to inspect one or more unassembled parts of the manufacturable entity, recognizing, using the input device, an unassembled part from the one or more unassembled parts, wherein the input device is coupled to an electronic device, determining a correct location and relationship of the recognized unassembled part in relation to one or more of the one or more unassembled parts in a pre-assembled structure, and displaying, on a display, an animation in an augmented reality computing environment demonstrating the relationship between the recognized unassembled part and the one or more unassembled parts, both in regard to each other and to the structure itself.

Abstract: Architecture that detects entrances on building facades. In a first stage, scene geometry is exploited and the multi-dimensional problem is reduced down to a one-dimensional (1D) problem. Entrance hypotheses are generated by considering pairs of locations along lines exhibiting strong gradients in the transverse direction. In a second stage, a rich set of discriminative image features for entrances is explored according to constructed designs, specifically focusing on properties such as symmetry and color consistency, for example. Classifiers (e.g., random forest) are utilized to perform automatic feature selection and entrance classification. In another stage, a joint model is formulated in three dimensions (3D) for entrances on a given facade, which enables the exploitation of physical constraints between different entrances on the same facade in a systematic manner to prune false positives, and thereby select an optimum set of entrances on a given facade.

Abstract: Systems and methods for time series prediction are described. The systems and methods include encoding driving series into encoded hidden states, the encoding including adaptively prioritizing driving series at each timestamp using input attention, the driving series including data sequences collected from sensors. The systems and methods further includes decoding the encoded hidden states to generate a predicting model, the decoding including adaptively prioritizing encoded hidden states using temporal attention. The systems and methods further include generating predictions of future events using the predicting model based on the data sequences. The systems and methods further include generating signals for initiating an action to devices based on the predictions.

Abstract: Systems and devices including an imaging sensor to capture video sequences in an environment having safety concerns therein. The systems and devices further including a processor to generate driving series based on observations from the video sequences, and generate predictions of future events based on the observations using a dual-stage attention-based recurrent neural network (DA-RNN). The DA-RNN includes an input attention mechanism to extract relevant driving series, an encoder to encode the extracted relevant driving series into hidden states, a temporal attention mechanism to extract relevant hidden states, and a decoder to decode the relevant hidden states. The processor further generates a signal for initiating an action to machines to mitigate harm to items.

Abstract: A first sequence of video fingerprints at a first image repetition rate is compared with a second sequence of video fingerprints at a second, different image repetition rate. A converted sequence of fingerprints is generated from the first or second sequence of fingerprints by forming a combination of the values of a number of neighbouring fingerprints. The combination is a weighted average based on the temporal location of the respective neighbouring fingerprints in the first sequence of fingerprints with respect to the fingerprint in the converted sequence. A correlation is then performed between the converted sequence of fingerprints and the other sequence of fingerprints.

Abstract: Flat panel displays can broadcast their identity for discovery by handheld devices interested in obtaining data associated wit, video programming displayed on the displays. An application can be downloaded and activated on smartphones that will enable them to capture an image of the video programming and recognize is with the help of a remote server, or wirelessly obtain a list of available (discovered) flat panel displays (multiple screens, depending on venue) that allow wireless access to data associated with video programming displayed (or previously displayed and stored in a queue) on flat panels. Video programming rendering on a flat panel of interest can be identified via the smartphone and the smartphone can access additional data either directly from the flat panel or from a remote server once the data has been identified and/or engage in an interactive session in association with the displayed programming or data.

Abstract: A method and system associated with a camera view of a moving-object in a scene. The method comprises detecting and tracking the moving object over multiple video frames, estimating an orientation of the moving object in each of the video frames, and constructing a cost map from the estimated orientations over the multiple video frames for finding a minimum cost path over the cost map. The Method also comprises determining regularised orientation estimates of the moving-object from the minimum cost path, and locating the vanishing point of the camera view based on an axis of the moving-object from the minimum cost path, the axis formed by using the regularised orientation estimates.

Abstract: An example method includes determining first classification rule(s) based on first data that classifies pixel groups of first reference image(s) as types of objects. The first reference image(s) are captured by a first sensor. The method further includes determining second classification rule(s) based on second data that classifies pixel groups of second reference image(s) as types of objects. The second reference image(s) are captured by a second sensor. The method includes classifying, as respective first objects, pixel group(s) of a first image of a scene based on the first classification rule(s). The method includes classifying, as respective second objects, pixel group(s) of a second image of the scene based on the second classification rule(s). The method further includes determining whether a change occurred in the scene based on the object classification of the first image(s) and the second image(s).

Abstract: An image processing device, an imaging device, an equipment control system, equipment, an image processing method, and a recording medium storing a program. The image processing device and the image processing method includes capturing an object to obtain a captured image using at least one capturing unit, generating a correction parameter based on the captured image to update a correction parameter stored in a memory, correcting displacement in coordinate caused to the captured image, using the updated correction parameter, and updating a referring status of a specific thing included in the captured image to a not-yet-referred status when the correction parameter is updated. The imaging device includes the image processing device, the at least one capturing unit captures an image including the specific thing, and the memory stores the correction parameter.

Abstract: A system that detects the number of persons waiting in a queue includes an image acquisition unit that acquires a captured image, an analysis unit that detects a person from the captured image, and a decision unit that, when a region in which no person is detected in the captured image has an area greater than or equal to a predetermined area, sets the number of persons detected from the captured image as the number of persons waiting in the queue.

Abstract: A computer implemented method for detecting an object in a crowded scene utilizing an image capturing device. The method includes receiving an image of a predetermined area. From the image, the existence of selected portions as representing an entity of a selected class is determined. Each selected portion is assigned an initial confidence value that the selected portion is an entity representative of a selected class. Each selected portion is evaluated with each other to determine a context confidence value. The context confidence value and initial confidence value are utilized to determine which of the one or more selected portions are entities of a selected class.

Abstract: A method for determining wetness on a path of travel. A surface of the path of travel is captured by at least one image capture device. A plurality of wet surface detection techniques is applied to the at least one image. An analysis for each wet surface detection technique is determined in real-time of whether the surface of the path of travel is wet. Each analysis independently determines whether the path of travel is wet. Each analysis by each wet surface detection technique is input to a fusion and decision making module. Each analysis determined by each wet surface detection technique is weighted within the fusion and decision making module by comprehensive analysis of weather information, geology information, and vehicle motions. A wet surface detection signal is provided to a control device. The control device applies the wet surface detection signal to mitigate the wet surface condition.

Abstract: A method and apparatus for controlling a vision sensor are provided. The apparatus and corresponding method are configured to predict an expected point, on a traveling path of a host vehicle, at which an illumination variation greater than or equal to a threshold is expected to occur, and determine whether the host vehicle is located within a threshold distance. The apparatus and corresponding method are also configured to control a vision sensor in the host vehicle based on the expected illumination variation in response to the host vehicle being located within the threshold distance.

Abstract: A method for detecting and evaluating environmental influences and road condition information in the surroundings of a vehicle. At least two digital images are generated in a successive manner using a camera, and the same image section is selected on each image. Changes in the image sharpness between the image sections of the at least two successive images are detected using digital image processing algorithms, wherein the image sharpness changes are weighted in a decreasing manner from the center of the image sections towards the outside. Surroundings condition information is ascertained on the basis of the detected image sharpness changes between the image sections of the at least two successive images using machine learning methods, and road condition information is determined on the basis of the ascertained surroundings condition information.

Abstract: A method for detecting a lane line is provided. The method includes: determining candidate lane line regions in a lane image by filtering the lane image; extracting center points for each candidate lane line region; obtaining a fitting curve of each candidate lane line region by performing curve fitting on the center points of each candidate lane line region; classifying the candidate lane line regions as rectilinear and curvilinear candidate lane line regions according to a slope angle of the fitting curve of each candidate lane line region; dividing the candidate lane line regions into groups based on the classifying result; and determining two groups, which meet a first predetermined condition, as target groups, and recognizing candidate lane line regions in the target groups as lane line regions. Besides, an apparatus and a device for detecting a lane line are further provided.

Abstract: In a vehicle-mounted image recognition apparatus, a resolution of an image projected on an imaging plane of an image sensor is different at a position away from a center between a circumferential direction and a radial direction. To make a circumferential resolution higher than a radial resolution, at least a lower half of the imaging plane is located closer to a circumferential focus than a middle of a radial focus and the circumferential focus at a position off the optical axis center of the image projected on the image sensor through the image-forming optical system. This is achieved by adjusting the position of the imaging plane when manufacturing or by selecting a lens with high circumferential resolution. This apparatus improves recognition accuracy in recognizing traffic lanes without using an expensive lens exhibiting high resolution both in the circumferential direction and in the radial direction.

Abstract: A vanishing point correction method may include the steps of: (a) extracting a contour candidate point of an object considered as the forward vehicle from the image taken by the camera; (b) tracking a movement of the contour candidate point over time; (c) determining whether the contour candidate point belongs to the vehicle; (d) setting the contour candidate point to a vehicle contour point of the forward vehicle, when it is determined at the step (c) that the contour candidate point belongs to the vehicle; (e) calculating a variation in width between vehicle contour points at the same height among the vehicle contour points; and (f) calculating a variation of the vanishing point.

Abstract: Aspects may relate to a device to provide a spoofing or no spoofing indication. The device may comprise a processor and a sensor. The sensor may receive multiple facial frames of a face of a user. The processor coupled to the sensor may be configured to: perform a function based upon components of the face of the user relative to one another to determine measured facial features for two adjacent frames of the multiple facial frames; and determine whether the measured facial features for the two adjacent frames are sufficiently different to indicate liveness of the user and no spoofing attempt.

Abstract: Methods, systems, and computer program products for authenticating an online user. Authentication involves sending a code from a server to a user device equipped with a source of illumination and a camera capable of capturing video imagery of the online user. The user device receives the code, modulates the source of illumination in accordance with the code, and captures video imagery of the user while the source of illumination is being modulated according to the code. The captured video imagery of the online user is sent to the server where it is analyzed to detect evidence of changes in illumination that correspond to the code. If good correspondence is found, the user may be authenticated. Similar methods may be applied to other biometric data. Applications of the authentication include identify validation, pseudonym verification, and distinguishing human from non human access attempts.

Abstract: Provided is a parallax minimization stitching method and apparatus using control points in an overlapping region. A parallax minimization stitching method may include defining a plurality of control points in an overlapping region of a first image and a second image received from a plurality of cameras, performing a first geometric correction by applying a homography to the control points, defining a plurality of patches based on the control points, and performing a second geometric correction by mapping the patches.

Abstract: A computing platform is described to match a palm digital representation to a palm template. The platform includes logic causing presenting a viewfinder image including an image stream and/or a cue appearing to reside substantially in front of a user to guide the user to align a palm at a position within a field of view of a camera; capturing a set of images using the camera; processing the set of images to determine a set of identifying features of the palm; and comparing the set of identified features with enrolled palm prints to identify the palm.

Abstract: The present application discloses a vehicle searching method and device, which can perform the steps of: calculating an appearance similarity distance between a first image of a target vehicle and several second images containing the searched vehicle; selecting several images from the several second images as several third images; obtaining corresponding license plate features of license plate areas in the first image and each of the third images with a preset Siamese neural network model; calculating a license plate feature similarity distance between the first image and each of the third images according to license plate feature; calculating a visual similarity distance between the first image and each of the third images according to the appearance similarity distance and the license plate feature similarity distance; obtaining a the first search result of the target vehicle by arranging the several third images in an ascending order of the visual similarity distances.

Abstract: A view friendly monitor system provides a monitor and a sequence of views that are regions of interest. The monitor can display a region of the camera view; the system is configured such that the displayed region of the camera view is a region of interest to the user. The displayed portion of the camera view changes to match the changing interests of the user. For example, a user that is interested to see a view in a direction beyond a certain edge of the monitor would move her head generally in an opposite direction. To see farther outside the edge, she needs to move her head farther in the opposite direction. In response to her head movement, the view friendly monitor system would change the region of interest based on the direction and the length of her head movement and display the new region of interest on the monitor.

Abstract: The invention relates to a method for OCR detection of valuable documents in a cash dispenser in the case of which an image of the valuable document is detected by means of a digital video or matrix camera. A Hough transformation is used to calculate edge lines of the valuable document and a rotation angle is calculated therefrom such that the edges of the valuable document are aligned with the image edges. The detected image is homogenized to compensate an inhomogeneous image background. This is followed by OCR detection of alphanumeric information on the valuable document.

Abstract: The present disclosure discloses an image recognition method and system based on deep learning. The image recognition method trains a recognizing engine for high resolution images and a recognizing engine for low resolution image separately. The two recognizing engines generate two independent feature groups specifically for high resolution and low resolution images respectively. A categorizing engine categorizes a testing image to a high resolution mode or a low resolution mode, and then the testing image is recognized by the appropriate recognizing engine. Therefore, the present disclosure increases the accuracy of image recognition under various conditions.

Abstract: Techniques for locating features of interest in warped images are described. For example, a destination image is generated from a source image based on warping a portion of the source image according to a backward map. Each of the two images is associated with a mesh having a tile-based geometry. The backward map maps points from the destination mesh to pixels in the source image. Triangles are formed from the tiles of the destination mesh. A forward map is generated based on the triangles and the backward map. The forward map maps points from the source mesh to pixels in the destination image. A point from the source mesh is identified that corresponds to a pixel of a feature of interest. The forward map is applied to this point to locate a pixel in the destination image. This pixel corresponds to the location of the feature in the destination image.

Abstract: A method for producing digital ink in a collaboration session between a first computing device and a second computing device that presents a digital canvas. In some embodiments, the method includes capturing a raster image of content using a camera operably coupled to the first computing device, deriving first image vectors and second image vectors based on first and second portions, respectively, of the raster image, sending the first image vectors to the second computing device for displaying a first digital ink object based on the first image vectors, and sending the second image vectors to the second computing device for displaying a second digital ink object based on the second image vectors after the displaying of the second digital ink object.

Abstract: Method and device for capturing images of a target object using image templates at a device are disclosed. The method includes rendering an image template on the display for determining a position of the target object in an image to be captured using the device; capturing a first image of the target object using the image sensor and displaying the captured first image on the display while keeping the image template on the display; extracting one or more features associated with the position of the target object from the captured first image; determining whether the one or more features extracted from the captured first image match the image template; and generating an alert signal in accordance with a determination that at least one of the one or more features extracted from the captured first image matches the image template.

Abstract: An imaging method includes obtaining an image with a first field of view and first effective resolution and the analyzing the image with a visual attention algorithm to one identify one or more areas of interest in the first field of view. A subsequent image is then obtained for each area of interest with a second field of view and a second effective resolution, the second field of view being smaller than the first field of view and the second effective resolution being greater than the first effective resolution.

Abstract: A method is provided for evaluating the authenticity of a painting which includes providing (10) a collection of a plurality of original paintings by a certain artist and creating (20) at least one digital image of each original painting from the collection of plurality of original paintings; detecting (30) classification data for each one of the digital images of each original painting of the collection using the WND-CHARM method, wherein the classification data a reference data record of classification data for original paintings by the artist; providing (40) an additional painting and creating (50) at least one digital image of the additional painting; detecting (60) additional classification data for the digital image of the additional painting using the WND-CHARM method and comparing (70) the additional classification data with the reference data record of classification data for original paintings, wherein on reaching or exceeding a minimum correspondence between the additional classification data and t

Abstract: An object recognition apparatus includes an image capturing unit, a storage unit that stores feature data of a plurality of reference objects, a display unit, and a processor. The processor calculates degrees of similarity between an object in the captured image and a plurality of reference objects, based on the feature data of the reference objects, selects some of the reference objects as candidates objects based on the calculated degrees of similarity, and controls the display unit to display the candidate objects and a message to select one of the candidate objects. If no candidate object is selected within a predetermined time after the candidate objects and the message have been displayed, and thereafter a new image is captured by the image capturing unit, the processor removes the message and calculates degrees of similarity between an object in the newly captured image and the plurality of reference objects.

Abstract: A similarity detection system receiving a plurality of input entities, the system including a cohesive subgraph identification device configured to calculate, based on attributes of the plurality of input entities, a first parameter and a second parameter based on the first parameter, and further configured to identify a plurality of subgraphs from the second parameter and a subgraph correlation tracking and clustering device configured to determine a relationship between different subgraphs based on a similarity factor between the second parameter and the plurality of subgraphs.

Abstract: A face clustering system for video face clustering in a video sequence, the system including an inherent supervision summarization device configured to collect group-level supervision and instance level supervision within a same chunklet based on a user input of face images for a person, a discriminative projection learning device configured to embed group constraints of the group-level supervision into a transformed space, and configured to generate an embedding space from the original image feature space, and a clustering device, in the embedding space, configured to execute pair-wise based clustering to cluster the video images into different clusters with the instance level supervision collected by the inherent supervision summarization device.

Abstract: A clustering and recommendation machine determines that an item is included in a cluster of items. The machine accesses item data descriptive of the item. The machine accesses a vector that represents the cluster and calculates the likelihood that the item is included in the cluster, based on the item variable and the probability parameter. The machine determines that the item is included in the cluster, based on the likelihood. The machine also recommends an item to a potential buyer. The machine accesses behavior data that represents a first event type pertinent to a first cluster of items. The machine calculates a probability that a second event type pertaining to a second cluster of items will co-occur with the first event type. The machine identifies an item from the second cluster to be recommended and presents a recommendation of the item to the potential buyer.

Abstract: Provided is an object detection device for efficiently and simply selecting an image for creating instructor data on the basis of the number of detected objects. The object detection device is provided with: a detection unit for detecting an object from each of a plurality of input images using a dictionary; an acceptance unit for displaying, on a display device, a graph indicating the relationship between the input images and the number of subregions in which the objects are detected, and displaying, on the display device, in order to create instructor data, one input image among the plurality of input images in accordance with a position on the graph accepted by operation of an input device; a generation unit for generating the instructor data from the input image; and a learning unit for learning a dictionary from the instructor data.

Abstract: In a video frame processing system, a feature extractor generates, based on a plurality of data sets corresponding to a plurality of frames of a video, a plurality of feature sets, respective ones of the feature sets including features extracted from respective ones of the data sets. A first stage of the feature aggregator generates a kernel for a second stage of the feature aggregator. The kernel is adapted to content of the feature sets so as to emphasize desirable ones of the feature sets and deemphasize undesirable ones of the feature sets. In the second stage of the feature aggregator the kernel generated by the first stage is applied to the plurality of feature sets to generate a plurality of significances corresponding to the plurality of feature sets. The feature sets are weighted based on corresponding significances and weighted feature sets are aggregated to generate an aggregated feature set.