Abstract: The present invention relates to a method for assisting multiple users to perform a collection simultaneously. The method includes the steps of: (a) acquiring digital data created with respect to recognition reference information of an object from a terminal of each of the multiple users; (b) determining or recognizing whether the respective digital data on the recognition reference information acquired through the terminals were created within a preset place condition and whether the respective digital data on the recognition reference information acquired through the terminals were created within a preset scope of the time; (c) selecting a specified group of users, including a first to an n-th user among the multiple users, who create the digital data within the preset place condition and within the preset scope of the time; and (d) providing information on rewards corresponding to the object for users included in the specified group of users.

Abstract: The present invention relates to a method for providing advertisement by using collection information. The method includes the steps of: allowing recognition reference information on an object from a vendor server to be received; allowing collection information to be acquired; allowing the object corresponding to the basic recognition information to be recognized by comparing between the basic recognition information included in the digital data in the collection information acquired from the user terminal and the recognition reference information received from the vendor server and allowing statistical collection information to be calculated by referring to the information on the recognized object and a user of the user terminal who remits the basic recognition information; and allowing advertising information on a peculiar object to be provided for a user terminal(s) included in a specific user group.

Abstract: The present invention relates to a method for making a user easily replay a scene on which a specific person appears included in a movie. The method includes the steps of: (a) distinguishing persons appearing in the movie by recognizing their faces; (b) obtaining identifier of each distinguished person, feature data of the persons, information on time zone in which each person appears, wherein the information on the time zone includes a starting time and an ending time and wherein the ending time for a person is determined as the last time that the person appeared if the person has not appeared for more than a certain critical time; (c) displaying the identifier at a request of replay of the movie; and (d) replaying the movie from the starting time for a specific person who corresponds to a specific identifier if the specific identifier is selected.

Abstract: The present invention relates to a method for generating a panoramic image. The method includes the steps of: (a) adjusting resolutions of a first and a second input images, respectively, to thereby generate a first and a second adjusted images, wherein the resolutions of the first and the second adjusted images are determined by referring to pre-fixed relationship data with respect to the resolutions of the adjusted images to those of the input images; (b) generating a first and a second pre-processed images which represent information on edges of the first and the second adjusted images by referring to tangent vector components vertical to gradient vector components; and (c) performing image matching operations between the first and the second pre-processed images and then determining an optimal overlapped position where the first and the second input images are synthesized by referring to results of the image matching operations.

Abstract: The present invention relates to a method for writing a newly recognized image. The method includes the steps of: (a) comparing pre-stored image in the image database with a queried image; (b) storing the queried image onto a database for unrecognized images if there is no image similar to the queried image; (c) grouping the images in the database for unrecognized images based on degrees of similarity thereamong; and (d) comparing, if a specific image and its tag information are inputted, the specific image with some images included in a specific set of images among the organized sets of the images, determining whether there is any image in the specific set of images which has a degree of similarity exceeding the pre-set value and allowing images determined to have degrees of similarity exceeding the pre-set value with the tag information to be automatically written onto the image database.

Abstract: The present invention relates to a method for supporting a collection of an object included in an image inputted through a terminal. The method includes the steps of: recognizing the identity of an object by using at least one of an object recognition technology, an optical character recognition technology, and a barcode recognition technology; getting a collection page including at least part of the information on an auto comment containing a phrase or sentence correctly combined under the grammar of a language by using the recognition information and the information on the image of the recognized object; allowing the collection page to be stored when a request for registration of the page is received; and providing a specific user with the information about a reward system.

Abstract: The present invention relates to a method for recommending other users who have similar preference with a user. The method includes the steps of: acquiring information on objects selected by a user and other users; acquiring information on topics; determining to which topic the user and the other users respectively correspond; evaluating a degree of user similarity by each other user, the degree of user similarity being relevant to sum of probabilities of the specific object appearing in the state of a topic while k-th other user and the user have the topic as a probability of the k-th other user and the user having selected the specific object, by referring to correspondence relationships; and recommending specific other users who have a high degree of user similarity, wherein the degree of user similarity is evaluated by multiplying the sum by a probability of the specific object appearing independently.

Abstract: The present invention relates to a method for supporting a collection of an object included in a created image. The method includes the steps of: (a) creating an image of an object; (b) automatically creating and providing a combined sentence correct under the grammar of a language for the object on a first area on a screen of the terminal by using at least part of recognition information on what an identity of the object is, a place where the image was created and a time when the image was created, and automatically getting and providing a thumbnail corresponding to the recognized object on a second area on the screen of the terminal; and (c) if a Collection button is selected, storing data provided on the first and the second areas onto a storage space, to thereby complete the collection of the object.

Abstract: The present invention relates to a method for deblurring a barcode image. The method includes the steps of: acquiring the terminal's n OTFs calculated by varying focal lengths for a subject; if a barcode image is inputted through the mobile terminal, fixing a specific value among NSRs as a value to be applied to an error metric of Wiener filtering, applying n PSFs calculated from the n OTFs to the error metric by a binary search algorithm and determining whether the result of the application to the error metric is not exceeding the pre-set threshold or not; and selecting a PSF value and an NSR value as values of Wiener filtering if the result of the application to the error metric is not exceeding the pre-set threshold and then performing the Wiener filtering to the inputted barcode image by using the selected PSF value and the selected NSR value.

Abstract: The present invention relates to a method for providing information on an object by using viewing frustums. The method includes the steps of: (a) specifying at least two viewing frustums whose vertexes are visual points of respective user terminals; and (b) calculating a degree of interest in the object by referring to the object commonly included in both a first viewing frustum whose vertex is a visual point of a first user terminal and a second one whose vertex is a visual point of a second user terminal.

Abstract: There are provided a method and an apparatus for acquiring information on products which are located on the body of a person appearing in image data. More particularly, a method and an apparatus for analyzing a posture of a person from image data, i.e., a face posture and a body posture of the person, displaying, according to information on the posture of the person, at least one of candidate regions in which products are attached to the body of the person in the image data on a screen of a terminal of a user, and, if the user selects any one of the candidate regions, providing the terminal with information on products same as or similar to an image of a specific product included in the selected candidate region.

Abstract: The present invention relates to a method and a system for filtering the harmful content (s) which includes a filter group for providing an optimized filter for each category; a matching engine for monitoring harmfulness of the content by matching it with existing lewd contents recorded in a pornographic content database and/or advertising contents recorded in an advertising content database; an interface part which provides a user with information on a degree of similarity between the inputted content and the harmful contents recorded in the above-mentioned database which is calculated through the matching process in order to increase the filtering accuracy; and information on a degree of harmfulness of the content calculated in the filter group; and information on a user who created or distributed the content. Accordingly, it is possible to filter adult contents or advertising contents with much higher accuracy by three harmful content blocking steps.

Abstract: The present invention relates to a method for providing information on an object included in a visual field of a terminal in a form of augmented reality (AR) by using an image inputted to the terminal and its relating information. The method includes the steps of: (a) specifying the visual field of the terminal corresponding to the inputted image by referring to at least one piece of information on a location, a displacement and a viewing angle of the terminal; and (b) acquiring a graphic element corresponding to the object, included in the visual field of the terminal, whose identity is recognized by using a technology for matching a building image and displaying the acquired graphic element with the inputted image in the form of the augmented reality by providing the graphic element on a location of the object displayed on a screen of the terminal.

Abstract: The present invention relates to a method for applying a trimming operation to image. The method includes the steps of: detecting person and object; calculating area of face region and area of detected object and calculating distance between center of face region and that of image and distance between center of detected object and that thereof; and applying trimming operation to the object if area of detected object is by over a first prefixed percent larger than that of face, and applying trimming operation to the object if distance between center of detected object and that of image is a second prefixed value or less and distance between center of face and that of image is by over a third prefixed value larger than distance between center of detected object and that thereof.

Abstract: The present invention relates to a method for adaptively tracking a subject. The method includes the steps of: comparing a first block which indicates a region corresponding to a specific subject in a first frame with at least one block included in a second frame and determining a specific block among at least one block in the second frame which has the highest degree of similarity to the first block as a second block which indicates a region corresponding to the specific subject in the second frame; and detecting the specific subject from at least part of the whole region in the second frame by using a subject detection technology, if the degree of similarity between the first block and the second block is less than a predetermined threshold value, and resetting the second block in the second frame based on a region corresponding to the detected specific subject.

Abstract: The present invention relates to a method for providing information on an object excluded in a visual field of a terminal in a form of augmented reality (AR) by using an image inputted to the terminal and information related thereto. The method includes the steps of: (a) specifying the visual field of the terminal corresponding to the inputted image by referring to at least one piece of information on a location, a displacement and a viewing angle of the terminal; (b) searching an object(s) excluded in the visual field of the terminal; and (c) displaying guiding information on the searched object(s) with the inputted image in a form of the augmented reality; wherein the visual field is specified by a viewing frustum whose vertex corresponds to the terminal.

Abstract: The present invention relates to a method for setting initial values of graph cuts) to distinguish between a character region and a background region in an image. The method includes the steps of: (a) detecting a face region from an inputted image; (b) determining reference character region patterns including respective face regions, among pre-stored multiple character region patterns, of which degrees of similarity to the detected face region exceed the predetermined value; and (c) granting an initial value for the background to a first region which is not covered by any of the reference character region patterns and an initial value for the character to a second region which is covered by all of the reference character region patterns if the reference character region patterns are aligned overlappedly on the inputted image based on the detected face region.

Abstract: The present invention relates to a method for hierarchically matching a building image. The method includes the steps of: matching a wall of a specific building in the building image inputted as a query with a wall(s) of a building(s) in at least one panoramic image by using a technology of matching a building's shape or repeated pattern; selecting a candidate panoramic image(s) which includes a building(s) recognized to have the same or similar wall to the specific building in the panoramic image(s) as a result of matching its wall with others; matching at least one local region, if containing a recognizable string or figure, in the specific building with local region(s) in the building(s) of the candidate panoramic image(s) by using a technology of recognizing a string or a figure; and determining top n panoramic image(s) as the result of matching the local region.

Abstract: The present invention relates to a method for providing augmented reality (AR) by using an image inputted to a terminal and information relating to the inputted image. The method includes the steps of: (a) acquiring recognition information on an object included in the image inputted through the terminal; (b) instructing to search detailed information on the recognized object and providing a tag accessible to the detailed information, if the searched detailed information is acquired, on a location of the object appearing on a screen of the terminal in a form of the augmented reality; and (c) displaying the detailed information corresponding to the tag, if the tag is selected, in the form of the augmented reality; wherein, at the step (b), the information on the location of the object is acquired by applying an image recognition process to the inputted image.

Abstract: The present invention relates to a method for adaptively performing an image matching process according to a situation. The method includes the steps of: (a) recognizing the situation of a subject included in an inputted image; and (b) selecting an image group including at least one image to be compared with the inputted image of the subject for the image matching process among multiple images stored on a database, as a qualified reference image group, according to the recognized situation; wherein the images included in the qualified reference image group are dynamically changed according to the recognized situation.