Abstract: A camera calibration method and medium and a 3-dimensional (3D) object reconstruction method and medium using the camera calibration method are provided. The camera calibration method includes setting a feature track by tracking and matching features in an input image sequence; estimating 3-dimensional (3D) points in relation to the features by initializing a structure and motion of the camera; estimating a pose of the camera by using the 3D points and refining the structure of the camera based on the estimated pose; and refining the pose of the camera.

Abstract: A modeling method, medium, and system. The modeling method may include specifying an object in a plurality of 2 dimensional (2D) images expressed from different view points, considering an edge and contents of each component of each 2D image, generating a 3 dimensional (3D) model of each specified object, and matching the generated 3D models considering relative locations of each generated 3D model. Accordingly, a realistic 3D model of an object expressed in an image can be accurately generated.

Abstract: A modeling method, medium, and system. The modeling method identifies a object within an image by detecting edges within the image, determines a complexity of the identified object based upon detected surface orientations of the identified object relative to a defined surface of the image, selectively, based upon the determined complexity, generates one or more surfaces for a 3D model by identifying one or more vanishing points for one or more corresponding surfaces, respectively, of the identified object and respectively analyzing the identified one or more vanishing points relative to respective determined points of the one or more corresponding surfaces of the image, and generates the 3D model by combining the one or more surfaces in a 3D space.

Abstract: An apparatus and method of providing a panoramic image, the method including: preparing images of a plurality of levels, which are made by scaling each of two source images at several rates; computing a Sum of Square Difference (SSD) between two corresponding images in each of the plurality of levels of the two source images to form an image stitched together; and when stitching the two images, the overlapping area is divided into a plurality of parts and each part is assigned a weight which is linearly applied respectively to the divided part of each source image for blending color. The apparatus includes stitching units to stitch source images together at the plurality of levels and blending units to color blend the source images in the overlapped area.

Abstract: A method and apparatus of correcting geometric information and providing a panoramic view. In the method of providing a panoramic view, a tilting vector (X?, Y?) is generated by applying matrices H(a,b) and HF to a source image vector (X,Y). The matrix H(a,b) is defined by distortion angles “b” towards a plane direction and a distortion angle “a” towards a depth direction of the source image, and the matrix HF is defined according to a camera parameter matrix K. Also, a two-dimensional vector (x,y) of a source image is converted into a three-dimensional vector (X,Y,W) on a retinal image plane having a unit radial distance from an origin of a ground plane. Also, a warped image width (WIW) is estimated from a width and a field of view (FOV) of the source image. A warping vector (x?,y?) is generated based on the converted vector (X,Y,W) and the WIW.

Abstract: An apparatus and method for image-based 3D photorealistic head modeling are provided. The method for creating a 3D photorealistic head model includes: detecting frontal and profile features in input frontal and profile images; generating a 3D head model by fitting a 3D genetic model using the detected facial features; generating a realistic texture from the input frontal and profile images; and mapping the texture onto the 3D head model. In the apparatus and method, data obtained using a relatively cheap device, such as a digital camera, can be processed in an automated manner, and satisfactory results can be obtained even from imperfect input data. In other words, facial features can be extracted in an automated manner, and a robust “human-quality” face analysis algorithm is used.

Abstract: A modeling method, medium, and system. The modeling method may include specifying an object in a plurality of 2 dimensional (2D) images expressed from different view points, considering an edge and contents of each component of each 2D image, generating a 3 dimensional (3D) model of each specified object, and matching the generated 3D models considering relative locations of each generated 3D model. Accordingly, a realistic 3D model of an object expressed in an image can be accurately generated.

Abstract: A modeling method, medium, and system. The modeling method identifies a object within an image by detecting edges within the image, determines a complexity of the identified object based upon detected surface orientations of the identified object relative to a defined surface of the image, selectively, based upon the determined complexity, generates one or more surfaces for a 3D model by identifying one or more vanishing points for one or more corresponding surfaces, respectively, of the identified object and respectively analyzing the identified one or more vanishing points relative to respective determined points of the one or more corresponding surfaces of the image, and generates the 3D model by combining the one or more surfaces in a 3D space.

Abstract: A camera calibration method and medium and a 3-dimensional (3D) object reconstruction method and medium using the camera calibration method are provided. The camera calibration method includes setting a feature track by tracking and matching features in an input image sequence; estimating 3-dimensional (3D) points in relation to the features by initializing a structure and motion of the camera; estimating a pose of the camera by using the 3D points and refining the structure of the camera based on the estimated pose; and refining the pose of the camera.

Abstract: A method and apparatus of correcting geometric information and providing a panoramic view. In the method of providing a panoramic view, a tilting vector (X?, Y?) is generated by applying matrices H(a,b) and HF to a source image vector (X,Y). The matrix H(a,b) is defined by distortion angles “b” towards a plane direction and a distortion angle “a” towards a depth direction of the source image, and the matrix HF is defined according to a camera parameter matrix K. Also, a two-dimensional vector (x,y) of a source image is converted into a three-dimensional vector (X,Y,W) on a retinal image plane having a unit radial distance from an origin of a ground plane. Also, a warped image width (WIW) is estimated from a width and a field of view (FOV) of the source image. A warping vector (x?,y?) is generated based on the converted vector (X,Y,W) and the WIW.

Abstract: An apparatus and method of providing a panoramic image, the method including: preparing images of a plurality of levels, which are made by scaling each of two source images at several rates; computing a Sum of Square Difference (SSD) between two corresponding images in each of the plurality of levels of the two source images to form an image stitched together; and when stitching the two images, the overlapping area is divided into a plurality of parts and each part is assigned a weight which is linearly applied respectively to the divided part of each source image for blending color. The apparatus includes stitching units to stitch source images together at the plurality of levels and blending units to color blend the source images in the overlapped area.

Abstract: An apparatus and method for image-based 3D photorealistic head modeling are provided. The method for creating a 3D photorealistic head model includes: detecting frontal and profile features in input frontal and profile images; generating a 3D head model by fitting a 3D genetic model using the detected facial features; generating a realistic texture from the input frontal and profile images; and mapping the texture onto the 3D head model. In the apparatus and method, data obtained using a relatively cheap device, such as a digital camera, can be processed in an automated manner, and satisfactory results can be obtained even from imperfect input data. In other words, facial features can be extracted in an automated manner, and a robust “human-quality” face analysis algorithm is used.