Abstract: A method for spatial sound reproduction from a first audio input signal includes using a plurality of loudspeakers using a plurality of virtual loudspeakers over which the first audio input signal is panned for forming second audio input signals using source positioning data and virtual loudspeakers positioning data. The virtual loudspeakers are synthesized by modifying second audio input signals forming third audio input signals using virtual loudspeaker spatial filter coefficients that align the loudspeakers.

Abstract: An audio output control method and apparatus using a wired/wireless communication system is provided. An audio output control method of an audio output device connected to at least one other audio output device according to the present disclosure includes converting specific audio data to at least one modified audio signal designated to the respective audio output devices, playing at least one of the modified audio signals through a channel designated to the specific audio data, and transmitting a remaining of the modified audio signals to the at least one other audio output device, wherein the modified audio signals include at least one of sound sources and channel characteristics of the specific audio data.

Abstract: A method for determining an audio precompensation controller for an associated sound generating system comprising a total of N?2 loudspeakers, each having a loudspeaker input. The audio precompensation controller has a number L?2 inputs for L input signal(s) and N outputs for N controller output signals, one to each loudspeaker. It is relevant to: estimate (S1), for each one of at least a subset of the N loudspeaker inputs, an impulse response at each measurement position; specify (S2), for each one of the L input signal(s), a selected one of the N loudspeakers as a primary loudspeaker and optionally a selected subset S including at least one of the N loudspeakers as support loudspeaker(s); select (S2) at least one loudspeaker pair, that is required to be symmetrical with respect to the listing position; and specify (S3), for each primary loudspeaker, a target impulse response at each measurement position.

Abstract: An electronic device having hearing aid capability is disclosed. The electronic device includes a linear resonance actuator (LRA). The LRA is configured to produce both vibrations and a hearing aid compatible magnetic field. A processor coupled to the LRA is operable to control the LRA to generate a magnetic field for inductively coupling the electronic device to a hearing aid when a hearing aid compatibility (HAC) mode is enabled, wherein the generated magnetic field is encoded with audio information which corresponds to an audio signal.

Abstract: A hearing aid with an assembly, the assembly includes: a first side; a second side; a signal processor; a wireless communications unit, the wireless communications unit being connected to the signal processor; and an antenna for electromagnetic field emission and electromagnetic field reception, the antenna being connected to the wireless communications unit, the antenna having an excitation point; wherein a first branch of the antenna extends from the excitation point and a second branch of the antenna extends from the excitation point, at least a part of the second branch extending from the first side to the second side, and wherein the second branch has at least one ground connection.

Abstract: A hearing aid includes: a signal processor; a wireless communications unit, the wireless communications unit being connected to the signal processor; and an antenna for electromagnetic field emission and electromagnetic field reception, the antenna being connected to the wireless communications unit, the antenna having an excitation point; wherein a first branch of the antenna extends from the excitation point and a second branch of the antenna extends from the excitation point; and wherein the antenna has a first resonant frequency and a second resonant frequency.

Abstract: A hearing aid with an assembly, the assembly includes: a microphone; a signal processor for processing a first audio signal into a second audio signal compensating a hearing loss of a user; a wireless communication unit configured for wireless data communication; and an antenna for emission of an electromagnetic field, the antenna being coupled with the wireless communication unit, the antenna having a total length between three quarters of a wavelength of the emitted electromagnetic field and five quarters of the wavelength; wherein a part of the antenna extends from a first side of the assembly to a second side of the assembly; and wherein the antenna has a midpoint located at the part of the antenna extending from the first side to the second side, or a distance between the midpoint of the antenna and the part of the antenna is less than a quarter of the wavelength.

Abstract: A hearing aid with an assembly, the assembly includes: a first side; a second side; a signal processor; a wireless communications unit, the wireless communications unit being connected to the signal processor; and a monopole antenna for electromagnetic field emission and electromagnetic field reception, the antenna being connected to the wireless communications unit, the antenna having an excitation point; wherein at least a part of a first branch of the antenna extends from the excitation point along the first side of the assembly, the first branch having a first end, the first end being free or being coupled with the excitation point via a third branch; and wherein a second branch of the antenna extends from the excitation point, at least a part of the second branch extending in a space located between the first side of the assembly and the second side of the assembly.

Abstract: A hearing aid includes: a signal processor; a wireless communications unit connected to the signal processor; and an antenna system with a data terminal and at least one control terminal, wherein the data terminal is connected to the wireless communications unit, the antenna system comprising a first antenna and one or more antenna feeds including a first antenna feed, the antenna system further comprising an antenna configurator for configuring the antenna system to operate in a selected antenna mode selected from a plurality of antenna modes including a first antenna mode and a second antenna mode according to a control signal transmitted on the at least one control terminal, wherein the antenna configurator is configured to apply a first antenna configuration to the data terminal when operating in the first antenna mode, and to apply a second antenna configuration to the data terminal when operating in the second antenna mode.

Abstract: A battery assembly for powering a hearing device includes: a main part having at least a first surface and a second surface, the main part comprising a first electrode formed as an inductive coil; a first attachment member on at least a primary part of the first surface for attaching the main part to a hearing device housing of the hearing device; a second attachment member for attaching a direct current source to the main part; and an electronic circuit having a first input terminal and a second input terminal connectable to a first output terminal and a second output terminal, respectively, of the direct current source, wherein the electronic circuit is connected to the first electrode, and wherein the electronic circuit is configured for converting direct current from the direct current source to alternating current.

Abstract: A frame has an interior side and an exterior side. The frame is configured to extend around the back of a user's head. A first membrane is coupled to at least a portion of the interior side of the frame. In one embodiment, a second membrane is coupled to the first membrane. The first membrane and the second membrane define a receptacle and an opening that communicates with the receptacle. In one embodiment, a speaker is disposed in the receptacle. A first electrical wire has a first end electrically coupled to the speaker and a second end including a connector. The connector can be disposed proximate to the opening of the receptacle. The connector is configured to be electrically coupled to a second electrical wire.

Abstract: An easy assembled and detached touch includes a pen body having a hollow portion, a touch end, a first opening, a second opening; the hollow portion having an oblong shape and penetrating through a front and a rear side thereof; the first opening and second opening is opened at an upper side of the pen body and is communicated to the hollow portion; the touch end having a conductive head; a sound guiding unit being a protrusion having a size approximately matched to that of the hollow portion; a sound path for guiding sound from one end to another end of the sound guiding unit; in assembly state; the sound path having one end coupled to an electronic device; and an inserting unit formed at the lateral side of the sound guiding unit which can be inserted into an insertion hole at a lateral side of the audio device.

Abstract: It is possible to enable sound, such as music or words, to be caught clearly without raising the sound volume in particular in the use of a mobile terminal device. In a mobile terminal device provided with a speaker on a rear surface side, a sound guiding device that guides sound generated by the speaker in a front direction has a case that covers the speaker when attached to the mobile terminal device. A sound guiding section that guides sound generated by the speaker roughly in the front direction and a sounding port through which the guided sound is emit bed to the outside are provided at a position of the case corresponding to the speaker of the mobile terminal device when the case is attached to the mobile terminal device.

Abstract: An earbud cover is configured to be removably attached to, and at least partially cover, an audio earbud. The removable earbud cover includes at least one light guide configured to guide light from at least one optical emitter associated with the audio earbud to a region of a body of a subject wearing the audio earbud and/or guide light from the body of the subject to at least one optical detector associated with the audio earbud. The removable earbud cover may include at least one alignment member that facilitates stability of the audio earbud when inserted within an ear of the subject. The removable earbud cover may include at least one lens region, at least one light diffusion region, and/or at least one luminescence region in optical communication with the at least one light guide.

Abstract: According to an aspect, a sound outputting device includes a front unit, a first side unit, a second side unit, a sound transmitting portion, and a piezoelectric speaker. The first side unit is coupled to one end portion of the front unit. The second side unit is coupled to another end portion of the front unit. The sound transmitting portion is provided in the first side unit for transmitting a sound via cartilage conduction. The piezoelectric speaker vibrates the sound transmitting portion.

Abstract: A headset apparatus is provided, the apparatus comprises: a headband; two speakers and a cord; the apparatus comprises grooves and/or channels for allowing the cord to be wrapped around the apparatus during a stowage position.

Abstract: An actuation system for generating a physical effect, the system comprising at least one array of translating elements each constrained to travel alternately back and forth along a respective axis, toward first and second extreme positions respectively, in response to activation of first and second forces respectively; and a controller operative to use the first and second forces to selectably latch at least one subset of said translating elements into the first and second extreme positions respectively.

Abstract: A low-voltage electronic device receptacle may include a back wall, one or more side walls extending from the back wall, and a flange extending outward from the one or more side walls, wherein the back wall includes a forwardly-extending generally convex protrusion making up greater than 10% of the back wall. An enclosure for installing a speaker in a building structure having first and second support members and an outer layer may include a receptacle including a back wall, one or more side walls, and a flange extending outwardly from the one or more side walls, wherein the receptacle may be positioned between the first and second support members, the back wall includes a generally convex protrusion extending toward the outer layer, and the speaker may be mounted to the outer layer and a portion of the speaker may deflect at least a portion of the protrusion away from the outer layer.

Abstract: Provided is an electronic machine that is provided with a speaker and that enables a direction of the speaker to be appropriately and arbitrarily changed. The electronic machine includes a speaker unit, housings, and a rotating device. The speaker unit is supported by the housing. The rotating device includes a first rotating device configured to rotate the speaker unit with respect to the housings about a direction intersecting with an axis of the housings serving as a rotational axis, and a second rotating device configured to rotate the speaker unit with respect to the housings about the axis of the housings serving as a rotational axis.

Abstract: A support base of a hand-held electronic apparatus configured for placing a hand-held electronic apparatus having a speaker is provided. The support base comprises an accommodating body which is formed with a hollow casing, and the shape of the accommodating body corresponds to a portion of the hand-held electronic apparatus such that the hand-held electronic apparatus can be placed therein. The characteristics of the present invention are as follows: The accommodating body comprises a first sound broadcasting hole and a second sound broadcasting hole, and when the hand-held electronic apparatus is placed in the accommodating body, the speaker of the hand-held electronic apparatus corresponds to the first sound broadcasting hole such that the speaker of the hand-held electronic apparatus can broadcast sound through the second sound broadcasting hole.

Abstract: A speaker microphone design that prevents, eliminates, and/or minimizes speaker mechanical failure caused by fine metal particles in the ambient environment. The contemplated invention includes embodiments that utilize one or more barrier cover installed at a location between the speaker diaphragm and the grille. The barrier cover can be a thin shell of various shapes and sizes, such as a dome-shaped structure. In other embodiments, the barrier cover can be a thin membrane, such as a Bayer® film. Additionally or alternatively, the speaker is positioned further back away from the grille, to provide a clearance space of at least 10 mm between the diaphragm and the grille. This novel design of clearance is also effective in preventing, eliminating, and/or minimizing speaker mechanical failure caused by fine metal particles in the environment.

Abstract: A system for enhanced sports analytics and/or content creation includes: an object tracking system that generates coordinate data corresponding to object motion in a sports event; a data processing module that receives the coordinate data from the object tracking system, analyzes the coordinate data with an event recognition algorithm that identifies and characterizes events and outcomes of interest, and catalogs the data in accordance with the identified events and outcomes into event profile data; a database that receives and stores the event profile data generated by the data processing module; a user application that accesses the event profile data from the database; and at least one processing unit that executes instructions stored in at least one non-transitory medium to implement at least one of the object tracking system, the data processing module, or the user application.

Abstract: The disclosure relates generally to geographic-based signal detection. One claim recites an apparatus comprising: an input for receiving a signal from a cell phone; an electronic processor for determining, based at least in part on the signal, whether the cell phone is physically located in a predetermined home area; and upon a condition of not being in the predetermined home area, communicating a machine-readable code detector to the cell phone for use as its primary machine-readable code detector to detect machine-readable code while outside of its predetermined home area. Of course, other claims and combinations are provided as well.

Abstract: A method and system including receiving marking information, determining, at least in part, based on the marking 0 information, a plurality of color element additives, adding the plurality of color element additives to at least one color element of a video frame, wherein the at least one color element includes a color element R, a color element G, and a color element B. Related methods and systems are also described.

Abstract: An exemplary object detection method includes generating feature block components representing an image frame, and analyzing the image frame using the feature block components. For each feature block row of the image frame, feature block components associated with the feature block row are evaluated to determine a partial vector dot product for detector windows that overlap a portion of the image frame including the feature block row, such that each detector window has an associated group of partial vector dot products. The method can include determining a vector dot product associated with each detector window based on the associated group of partial vector dot products, and classifying an image frame portion corresponding with each detector window as an object or non-object based on the vector dot product. Each feature block component can be moved from external memory to internal memory once implementing the exemplary object detection method.

Abstract: A set of conditions associated with monitoring a given environment is received. One or more locations in the given environment are determined, based on at least a portion of the received set of conditions, for which data is to be acquired. The given environment is traversed through the one or more locations. Data at the one or more locations is acquired. The acquired data is stored for subsequent review. One or more of the above steps are performed under control of a processing device at least a part of which is mounted on a mobile platform that is configured to move through the given environment. Determination of the one or more locations in the given environment may further include determining an extended region to be observed.

Abstract: A computing device classifies user activities for a person interacting with a computer user interface using one or more user interface devices. The computing device receives eye tracking data for the person, which includes a sequence of fixations ordered temporally. Each fixation corresponds to a plurality of consecutive measured gaze points. Each fixation has a duration and location based on the corresponding gaze points. For each fixation, the computing device determines a plurality of features for the fixation, including characteristics of the fixation, context features based on preceding or subsequent fixations, and user interaction features based on information from the user interface devices during the fixation. The computing device assigns a user activity label to the fixation according to the features. The label is selected from a predefined set. The computing device then analyzes the fixations and their assigned user activity labels to make recommendations.

Abstract: A system and method for optimizing video-based tracking of an object of interest are provided. A video of a regularized motion environment that comprise multiple video frames is acquired and an initial instance of an object of interest in one of the frames is then detected. An expected size and orientation of the object of interest as a function of the location of the object is then determined. The location of the object of interest is then determined in a next subsequent frame using the expected size and orientation of the object of interest.

Abstract: In embodiments, apparatuses, methods and storage media for human-computer interaction are described. In embodiments, an apparatus may include one or more light sources and a camera. Through capture of images by the camera, the computing device may detect positions of objects of a user, within a three-dimensional (3-D) interaction region within which to track positions of the objects of the user. The apparatus may utilize multiple light sources, which may be disposed at different distances to the display and may illuminate the objects in a direction other than the image capture direction. The apparatus may selectively illuminate individual light sources to facilitate detection of the objects in the direction toward the display. The camera may also capture images in synchronization with the selective illumination. Other embodiments may be described and claimed.

Abstract: A stereo matching system comprising a face detection unit configured to detect a face area using either one of a reference image and a target image provided thereto to extract information about the detected face area and a support window setting unit configured to compare between the information about the detected face area by the face detection unit and a preset value to set the size of support window.

Abstract: A partial image extracting unit extracts images of a predetermined size and constant magnification from a tissue region. A mask generating unit generates a mask for removing a region not intended for measurement from the tissue region for each extracted image. A complete mask generating unit generates a temporary complete mask in which the masks generated for each of the images are integrated together, and generates a complete mask in which close portions among unmasked portions in the temporary complete mask are unified into one or more target regions. A measuring unit measures information pertaining to an object to be measured included in the image, and this information is measured for each of the images extracted by the partial image extracting unit. A region information calculating unit calculates, for each target region, information pertaining to the object to be measured from the measured information and from the complete mask.

Abstract: A gesture recognition device configured to detect a gesture from acquired image and generate command issued to a control target instrument according to the gesture, the gesture recognition device comprising: an image acquisition unit configured to acquire an image; a gesture acquisition unit configured to detect a target region performing a gesture from the acquired image, and acquire the gesture based on motion or a shape of the detected target region; a face detection unit configured to detect a face comprised in the acquired image; a correlation unit configured to correlate the detected target region with the detected face using a human body model representing a shape of a human body; a personal identification unit configured to identify a user corresponding to the detected face; and a command generation unit configured to generate a command issued to the control target instrument based on the identified user and the acquired gesture.

Abstract: A monitoring device configured to monitor whether a look of a target person is in a proper state suitable for a working environment, and comprising: an image input unit configured to input an image of the target person; a detector configured to analyze the image input by the image input unit and detect a predetermined region of a human body of the target person; a state estimator configured to estimate a state of the predetermined region detected by the detector; a proper state acquisition unit configured to acquire proper state information on a proper state of the predetermined region according to the working environment; and a controller configured to determine whether a present state of the predetermined region of the target person is proper by comparing an estimation result of the state estimator to the proper state information, and perform control according to a determination result.

Abstract: A vehicle is provided that is capable of preventing malfunction or inappropriate operation of the vehicle due to a passenger error by distinguishing a gesture of a driver from that of the passenger when a gesture of a user is recognized, and a method for controlling the same is provided. The vehicle includes an image capturing unit mounted inside the vehicle and configured to capture a gesture image of a gesture area including a gesture of a driver or a passenger. A controller is configured to detect an object of interest in the gesture image captured by the image capturing unit and determine whether the object of interest belongs to the driver. In addition, the controller is configured to recognize a gesture expressed by the object of interest and generate a control signal that corresponds to the gesture when the object of interest belongs to the driver.

Abstract: A non-transitory computer-readable medium storing a program for tracking a feature point in an image that causes a computer to execute a process, the process includes: calculating first values indicating degree of corner for respective pixels in another image, based on change of brightness values in horizontal direction and vertical direction; calculating second values indicating degree of similarity between respective areas in the another image and a reference area around the feature point in the image, based on comparison between the respective areas and the reference area; calculating third values indicating overall degree of corner and similarity, based on the first values and the second values; and tracking the feature point by identifying a point in the another image corresponding to the feature point in the image, based on the third values.

Abstract: A method and apparatus for tracking an object, and a method and apparatus for calculating object pose information are provided. The method of tracking the object obtains object feature point candidates by using a difference between pixel values of neighboring frames. A template matching process is performed in a predetermined region having the object feature point candidates as the center. Accordingly, it is possible to reduce a processing time needed for the template matching process. The method of tracking the object is robust in terms of sudden changes in lighting and partial occlusion. In addition, it is possible to track the object in real time. In addition, since the pose of the object, the pattern of the object, and the occlusion of the object are determined, detailed information on action patterns of the object can be obtained in real time.

Abstract: A system for tracking objects using an intelligent Video processing system in the context of airport surface monitoring. The system addresses airport surface monitoring operational issues such as all weather conditions, high robustness, and low false report rate. The output can be used to complement existing airport surface monitoring systems. By combining the use of multi-sensors and an adverse weather optimized system, the system is capable of producing an improved stream of information for the target object over traditional computer vision based airport surface monitoring systems.

Abstract: View-specific object detectors are learned as a function of scene geometry and object motion patterns. Motion directions are determined for object images extracted from a training dataset and collected from different camera scene viewpoints. The object images are categorized into clusters as a function of similarities of their determined motion directions, the object images in each cluster are acquired from the same camera scene viewpoint. Zenith angles are estimated for object image poses in the clusters relative to a position of a horizon in the cluster camera scene viewpoint, and azimuth angles of the poses as a function of a relation of the determined motion directions of the clustered images to the cluster camera scene viewpoint. Detectors are thus built for recognizing objects in input video, one for each of the clusters, and associated with the estimated zenith angles and azimuth angles of the poses of the respective clusters.

Abstract: An image such as a depth image of a scene may be received, observed, or captured by a device. A grid of voxels may then be generated based on the depth image such that the depth image may be downsampled. A background included in the grid of voxels may also be removed to isolate one or more voxels associated with a foreground object such as a human target. A location or position of one or more extremities of the isolated human target may be determined and a model may be adjusted based on the location or position of the one or more extremities.

Abstract: Technologies are generally described herein for extracting a representative motion flow from a video. Technologies are also generally described herein for retrieving a video utilizing the representative motion flow. The representative motion flow may be extracted utilizing a sliding window approach to generate interesting motion flows. The representative motion flow may be generated based on the interesting motion flows.

Abstract: Described embodiments include a system and method. A digital imaging device is configured to capture images of a region of a contact between a wheel of a terrestrial vehicle and a surface (“contact region”). A correlator is configured to correlate a first digital image of the contact region captured at a first time with a second digital image of the contact region captured at a second time. A kinematics circuit determines an incremental slide or slip of the wheel relative to the surface. A fraction status circuit combines at least two instances of the incremental slide or slip into data indicative of a slide or slip by the terrestrial vehicle relative to the surface. A communications circuit outputs an electronic signal indicative of the data indicative of a slide or slip by the terrestrial vehicle.

Abstract: An apparatus for measuring feet, for the purpose of gathering measurements for custom shoes, is disclosed. The apparatus is shaped like a shoe, and has an opening where the shoe lace opening is normally present on the shoe, in addition to openings near the big toe and the little toe. Each opening has holes for laces that are tightened so that the gap, or overlap, of the openings can be measured. Additionally, mats of known thickness can be used to measure the distance between the heel of the shoe and the heel of the foot. Furthermore, markings are used to identify the different shoe lace holes so measurements can be taken at multiple places along the top of the foot. Furthermore, a size indicator can be shown on the shoe, so that measurements can be taken on a picture of the foot in the shoe rather than on the foot and shoe themselves, and then scaled to actual size with reference to the size indicator.

Abstract: The invention is a method using a computer to construct a grid representative of the distribution of a physical property of an underground formation having application to petroleum reservoir development. A set of initial grid cells MI containing at least one conditioning point PC is defined. Each cell MI is then visited and assigned a value representative of the property using a filling method. At least one unvisited cell (MNV1, MNV2) adjacent to at least one already visited cell MV is then identified, and the number N of adjacent cells MV is determined for each cell (MNV1, MNV2). Each cell (MNV1, MNV2) is then visited and filled according to the decreasing values of N. The stages of identification and filling of cells (MNV1, MNV2) are repeated until each cell of the grid has been visited.

Abstract: A treatment recommending system of plant symptom includes an operation interface, an image capture unit, a processing unit and a storage unit for storing plant symptom data and diagnostic data. The processing unit may find target contour data presenting a plant part according to symptom description and symptom characteristic data. The image capture unit obtains plant image data corresponding to the target contour data. The processing unit analyzes the plant image data, and coordinates with the symptom characteristic data to determine corresponding treatment recommendation data.

Abstract: Systems and methods are described that may be used to match an image of an unknown animal, such as a lost pet, with images of animals that have been registered with an online service. Matching of the images of animals may be done in a two stage process. The first stage determines one or more images based on a classification of the images on the visual characteristics. The second stage determines a degree of matching between the retrieved images and the image to be matched.

Abstract: Processing method for transforming a first surface into a second surface, according to which: a/ in a (k+1)th iteration step, for each couple comprising a first elementary surface and a second elementary surface transformed from the first elementary surface in iteration k, the projection of the second elementary surface on the plane of the first elementary surface is determined and a matrix (A) is defined for said couple, where (A)=Pkt,(Pt)?1 and Pt=[pn+1?pi]i=1 to n, Pkt,=[p?n+1,k?p?i,k]i=1 to n; b/ the second elementary surfaces obtained for the (k+1)th iteration are determined as those minimizing a function including at least the term Eshape—k+1=(B).

Abstract: A method and system for allowing owners of a physical work, such as a book, to purchase an electronic version of the work by validating their ownership of the work using a personal electronic device, such as a smartphone, and marking the work. The owner is instructed to mark the work and to capture camera images of the work. The images are analyzed to identify the work, determine whether the work has been previously marked, and to verify that the work has been marked by the owner. Analysis may be performed remotely by a server in communication with the personal electronic device. Various means for authenticating ownership of the work are also disclosed, including 3D analysis, page curvature analysis, concurrent still image and video capture, detection of use of a video display to spoof the work, and geofencing.

Abstract: Methods, machine-readable media, and devices for floor plan space detection are described herein. For example, one or more embodiments include receiving an image of a floor plan, enhancing the image of the floor plan through processing, detecting a location of a door associated with a space in the image of the floor plan through a histogram of gradient of pattern feature, setting a seed in the space based on the location of the door associated with the space, spreading the seed in the space until the seed reaches a boundary of the space, and associating the seed that has been spread with the space.

Abstract: The present invention provides, in at least one aspect, methods and systems that detect at least one face in at least one digital image, determine and store area co-ordinates of a location of the at least one detected face in the at least one digital image, apply at least one transformation to the at least one detected face to create at least one portrait of the at least one detected face, rotate the at least one portrait at least until the at least one portrait is shown in a vertical orientation and a pair of eyes of the at least one face shown in the at least one portrait are positioned on a horizontal plane; and store the rotated at least one portrait.

Abstract: In an example embodiment, for each of the image exemplars, a first location offset between an actual landmark location for a first landmark in the image exemplar and a predicted landmark location for the first landmark in the image exemplar is determined. Then, a probability that the image recognition process applied using the first feature produces an accurate identification of the first landmark in the image exemplars is determined based on the first location offsets for each of the image exemplars. A weight may then be assigned to the first feature based on the derived probability. An image recognition process may then be performed on an image, the image recognition process utilizing a voting process, for each of one or more features, for one or more landmarks in the plurality of image exemplars, the voting process for the first feature weighted according to the weight assigned to the first feature.