Abstract: Disclosed herein are a markerless augmented reality system and method for extracting feature points within an image and providing augmented reality using a projective invariant of the feature points. The feature points are tracked in two images photographed while varying the position of an image unit, a set of feature points satisfying a plane projective invariant is obtained from the feature points, and augmented reality is provided based on the set of feature points. Accordingly, since the set of feature points satisfies the plane projective invariant even when the image unit is moved and functions as a marker, a separate marker is unnecessary. In addition, since augmented reality is provided based on the set of feature points, a total computation amount is decreased and augmented reality is more efficiently provided.

Abstract: Disclosed are a robot, which builds a map using a surface data of a three-dimensional image, from which a dynamic obstacle is removed, and a method of building a map for the robot. The method includes sequentially acquiring first and second surface data of a route on which the robot moves; matching the first and second surface data with each other to calculate a difference between the first and second surface data; detecting a dynamic obstacle from the first and second surface data according to the difference between the first and second surface data; generating a third surface data by removing the dynamic obstacle from at least one of the first and second surface data; and matching the third surface data and any one of the first and second surface data with each other to build the map.

Abstract: A 3D object tracking method and apparatus in which a model of an object to be tracked is divided into a plurality of polygonal planes and the object is tracked using texture data of the respective planes and geometric data between the respective planes to enable more precise tracking. The 3D object tracking method includes modeling the object to be tracked to generate a plurality of planes, and tracking the plurality of planes, respectively. The modeling of the object includes selecting points from among the plurality of planes, respectively, and calculating projective invariants using the selected points.

Abstract: A system and method for extracting 3D coordinates, the method includes obtaining, by a stereoscopic image photographing unit, two images of a target object, and obtaining 3D coordinates of the object on the basis of coordinates of each pixel of the two images, measuring, by a Time of Flight (TOF) sensor unit, a value of a distance to the object, and obtaining 3D coordinates of the object on the basis of the measured distance value, mapping pixel coordinates of each image to the 3D coordinates obtained through the TOF sensor unit, and calibrating the mapped result, determining whether each set of pixel coordinates and the distance value to the object measured through the TOF sensor unit are present, calculating a disparity value on the basis of the distance value or the pixel coordinates, and calculating 3D coordinates of the object on the basis of the calculated disparity value.

Abstract: Disclosed herein is a computer-readable medium and method of a mobile platform detecting and tracking dynamic objects in an environment having the dynamic objects. The mobile platform acquires a three-dimensional (3D) image using a time-of-flight (TOF) sensor, removes a floor plane from the acquired 3D image using a random sample consensus (RANSAC) algorithm, and individually separates objects from the 3D image. Movement of the respective separated objects is estimated using a joint probability data association filter (JPDAF).

Abstract: An augmented reality (AR) service apparatus includes a camera to capture an actual image, a controller to receive feature point information about the captured image from at least one of a plurality of base stations (BSs), detect a location of the camera by matching data of feature points with data of the image, and provide location-based information in a same direction as the captured image according to the location of the camera, and a display to realize an AR service by combining the captured image with the location-based information under control of the controller.

Abstract: A method and apparatus to build a 3-dimensional grid map and a method and apparatus to control an automatic traveling apparatus using the same. In building a 3-dimensional map to discern a current location and a peripheral environment of an unmanned vehicle or a mobile robot, 2-dimensional localization and 3-dimensional image restoration are appropriately used to accurately build the 3-dimensional grid map more rapidly.

Abstract: Disclosed herein are a method and apparatus for extracting feature points using hierarchical image segmentation and an image based localization method using the extracted feature points. An image is segmented using an affinity degree obtained using information observed during position estimation, new feature points are extracted from segmented areas in which registered feature points are not included, and position estimation is performed based on the new feature points. Accordingly, stable and reliable localization may be performed.

Abstract: An image processing apparatus for searching for a feature point by use of a depth image and a method thereof are provided. The image processing apparatus includes an input unit configured to input a three-dimensional image having depth information, a feature point extraction unit configured to obtain a designated point from an object image extracted from the depth image to obtain a feature point that is located at a substantially farthest distance from the designated point, and to obtain other feature points that are located at substantially farthest distances from feature points that are previously obtained as well as the designated point. The apparatus includes a control unit configured to control the input unit and the feature point extraction unit so that time in estimating a structure of the object is reduced, and a recognition result is enhanced.

Abstract: An apparatus and a method for pose recognition, the method for pose recognition including generating a model of a human body in a virtual space, predicting a next pose of the model of the human body based on a state vector having an angle and an angular velocity of each part of the human body as a state variable, predicting a depth image about the predicted pose, and recognizing a pose of a human in a depth image captured in practice, based on a similarity between the predicted depth image and the depth image captured in practice, wherein the next pose is predicted based on the state vector having an angular velocity as a state variable, thereby reducing the number of pose samples to be generated and improving the pose recognition speed.

Abstract: An image-based localization feature point registration apparatus includes a camera to capture an image, a feature point extractor to extract a feature point from the captured image, a calculator to calculate depth information about the feature point according to whether the feature point is one of two-dimensional (2D) and a three-dimensional (3D) corner, and a feature point register to register 3D coordinates of the feature point based on the depth information about the feature point and image coordinates of the feature point.

Abstract: A volume cell (VOXEL) map generation apparatus includes an inertia measurement unit to calculate inertia information by calculating inertia of a volume cell (VOXEL) map generator, a Time of Flight (TOF) camera to capture an image of an object, thereby generating a depth image of the object and a black-and-white image of the object, an estimation unit to calculate position and posture information of the VOXEL map generator by performing an Iterative Closest Point (ICP) algorithm on the basis of the depth image of the object, and to recursively estimate a position and posture of the VOXEL map generator on the basis of VOXEL map generator inertia information calculated by the inertia measurement unit and VOXEL map generator position and posture information calculated by the ICP algorithm, and a grid map construction unit to configure a grid map based on the recursively estimated VOXEL map generator position and posture.

Abstract: Two-dimensional image information and three-dimensional image information of a subject are acquired, facial recognition is performed using the two-dimensional image information to determine whether a recognized face is a registered user's face, an elliptical model of the user is matched to the three-dimensional image information to calculate an error if it is determined that the recognized face is the user's face, and it is determined whether the user's face is improperly used based on the error. The subject's face is determined using the two-dimensional image information and the three-dimensional image information of the subject and it is determined whether the recognized face is improperly used, thereby improving facial recognition reliability. Thus, information security is improved.

Abstract: Disclosed herein is a method of building a map of a mobile platform moving in a dynamic environment and detecting an object using a 3D camera sensor, e.g., an IR TOF camera sensor, for localization. A localization technology to separate and map a dynamic object and a static object is applied to a mobile platform, such as an unmanned vehicle or a mobile robot. Consequently, the present method is capable of accurately building map information based on the static object in a dynamic environment having a large number of dynamic objects and achieving a dynamic object avoidance or chasing function using position information acquired to build the map.

Abstract: An apparatus and a method of locating a moving robot are disclosed. The apparatus includes a storage unit storing information on straight lines of wall on a map, a state quantity detection unit detecting quantity of state of the robot running along the wall, and a control unit estimating an interior position of the robot by obtaining straight line information based on the detected state quantity and matching the obtained straight line information with the stored straight line information.

Abstract: A robot and a method for creating a robot path. The method for planning the robot path includes generating a depth map including a plurality of cells by measuring a distance to an object, dividing a boundary among the plurality of cells into a plurality of partitions according to individual depth values of the cells, and extracting a single closed loop formed by the divided boundary, obtaining a position and shape of the object located at a first time through the extracted single closed loop, calculating a probability that the object is located at a second time after t seconds on the basis of the obtained position and shape of the object located at the first time, and creating a moving path simultaneously while avoiding the object according to the calculated probability, thereby creating an optimum path without colliding with the object.

Abstract: A walking robot and a simultaneous localization and mapping method thereof in which odometry data acquired during movement of the walking robot are applied to image-based SLAM technology so as to improve accuracy and convergence of localization of the walking robot. The simultaneous localization and mapping method includes acquiring image data of a space about which the walking robot walks and rotational angle data of rotary joints relating to walking of the walking robot, calculating odometry data using kinematic data of respective links constituting the walking robot and the rotational angle data, and localizing the walking robot and mapping the space about which the walking robot walks using the image data and the odometry data.

Abstract: A system and method for extracting 3D coordinates, the method includes obtaining, by a stereoscopic image photographing unit, two images of a target object, and obtaining 3D coordinates of the object on the basis of coordinates of each pixel of the two images, measuring, by a Time of Flight (TOF) sensor unit, a value of a distance to the object, and obtaining 3D coordinates of the object on the basis of the measured distance value, mapping pixel coordinates of each image to the 3D coordinates obtained through the TOF sensor unit, and calibrating the mapped result, determining whether each set of pixel coordinates and the distance value to the object measured through the TOF sensor unit are present, calculating a disparity value on the basis of the distance value or the pixel coordinates, and calculating 3D coordinates of the object on the basis of the calculated disparity value.

Abstract: A volume cell (VOXEL) map generation apparatus includes an inertia measurement unit to calculate inertia information by calculating inertia of a volume cell (VOXEL) map generator, a Time of Flight (TOF) camera to capture an image of an object, thereby generating a depth image of the object and a black-and-white image of the object, an estimation unit to calculate position and posture information of the VOXEL map generator by performing an Iterative Closest Point (ICP) algorithm on the basis of the depth image of the object, and to recursively estimate a position and posture of the VOXEL map generator on the basis of VOXEL map generator inertia information calculated by the inertia measurement unit and VOXEL map generator position and posture information calculated by the ICP algorithm, and a grid map construction unit to configure a grid map based on the recursively estimated VOXEL map generator position and posture.

Abstract: A simultaneous localization and map building method of a mobile robot including an omni-directional camera. The method includes acquiring an omni-directional image from the omni-directional camera, dividing the obtained omni-directional image into upper and lower images according to a preset reference to generate a first image, which is the lower image, and a second image, which is the upper image, extracting feature points from the first image and calculating visual odometry information calculating visual odometry information to track locations of the extracted feature points based on a location of the omni-directional camera, and performing localization and map building of the mobile robot using the calculated visual odometry information and the second image as an input of an extended Kalman filter.