Abstract: Provided are a device and a method for detecting a wildfire. The wildfire detecting device may include: an image data acquisition module acquiring image data captured by a shooting device; an image data storage module storing the image data acquired through the image data acquisition module in a storage space; a wildfire detection module performing wildfire detection by using a deformable convolution based deep learning model to which smoke characteristics widely diffused are reflected by dynamically adjusting a filter position by using a learnable offset from the image data; and a wildfire direction and position estimation module estimating a direction and a position of the wildfire with respect to a case where it is determined that the wildfire occurs by the wildfire detection module.

Abstract: The embodiments disclosed herein relate to a biometric information registration apparatus and method using an artificial neural network. According to an embodiment, there are disclosed a biometric information registration apparatus and method that register biometric information by deriving converted information and a biometric information template based on an artificial neural network formed by successively connecting an encoding network for extracting security information and a feature point extraction network for extracting feature points to each other and renew the biometric information template using the converted information when updating the artificial neural network.

Abstract: A method for performing biometric recognition in a system in which biometric information is distributed and stored in a plurality of databases includes extracting biometric information of a user, generating a plurality of biometric information segments by dividing the extracted biometric information, calculating distances between pre-stored biometric information template segments for each of the plurality of databases and a corresponding one of the plurality of biometric information segments, and detecting a biometric information template matching the biometric information by using the calculated distances.

Abstract: Disclosed herein is a person re-identification method of identifying the same person from images taken through a plurality of cameras. The person re-identification method includes: detecting a person from an image taken by any one of a plurality of cameras; extracting bodily and movement path features of the detected person, and also extracting a facial feature of the detected person if the detection of the face of the detected person is possible; and matching the detected person for the same person against persons included in images taken by the plurality of cameras based on at least one of the bodily and facial features while reflecting a weight according to the movement path feature.

Abstract: A method for performing biometric recognition in a system in which biometric information is distributed and stored in a plurality of databases includes extracting biometric information of a user, generating a plurality of biometric information segments by dividing the extracted biometric information, calculating distances between pre-stored biometric information template segments for each of the plurality of databases and a corresponding one of the plurality of biometric information segments, and detecting a biometric information template matching the biometric information by using the calculated distances.

Abstract: The embodiments disclosed herein relate to a biometric information registration apparatus and method using an artificial neural network. According to an embodiment, there are disclosed a biometric information registration apparatus and method that register biometric information by deriving converted information and a biometric information template based on an artificial neural network formed by successively connecting an encoding network for extracting security information and a feature point extraction network for extracting feature points to each other and renew the biometric information template using the converted information when updating the artificial neural network.

Abstract: Disclosed herein is a method for creating augmented reality content. The method for creating augmented reality content includes: extracting a target area, to which augmented reality content will be applied, from an object included in an image; providing a template for the creation of the augmented reality content that will be applied to the extracted target area; when a user inputs user content via the template, creating the augmented reality content by using the user content and the metadata of the target area; and performing rendering so that the created augmented reality content is applied to the target area of the image.

Abstract: Disclosed herein is a method for creating augmented reality content. The method for creating augmented reality content includes: extracting a target area, to which augmented reality content will be applied, from an object included in an image; providing a template for the creation of the augmented reality content that will be applied to the extracted target area; when a user inputs user content via the template, creating the augmented reality content by using the user content and the metadata of the target area; and performing rendering so that the created augmented reality content is applied to the target area of the image.

Abstract: The present invention relates to an articulated model registration method and apparatus which three-dimensionally restore an articulated model of a body, such as a hand, an upper body, and a whole body, from an image and to acquire a two-dimensional landmark from a two-dimensional image to generate and estimate a three-dimensional articulated model corresponding thereto.

Abstract: The present invention relates to an articulated model registration method and apparatus which three-dimensionally restore an articulated model of a body, such as a hand, an upper body, and a whole body, from an image and to acquire a two-dimensional landmark from a two-dimensional image to generate and estimate a three-dimensional articulated model corresponding thereto.