Abstract:
PURPOSE: An automatic rigging and form transition method through the parametric control of a muscle and exterior curved surface-based object standard mesh model is provided to offer a 3D mesh model capable of free form transform and animation. CONSTITUTION: The multi-view image information of a dynamic object is collected(S101). The geometrical form of a 3D exterior is restored on the basis of volume using voxel(S102). A hierarchical joint structure is generated in a volume model(S103). The hierarchical joint structure keeps the characteristic of the hierarchical skeleton structure of a standard mesh model. Globally, the standard mesh model is scaled in a joint unit(S104).
Abstract:
본 발명은 개체 표준 메쉬 모델의 파라메트릭 제어를 통한 자동 리깅 및 형상 전이 방법에 관한 것으로, 골격구조를 갖는 동적 개체의 스캔 데이터 또는 입력된 삼차원 메쉬(Mesh) 모델로부터 조인트-스켈레톤(Joint-Skeleton) 구조를 추출하고, 해부학 정보에 근거하여 근육 및 외형 곡면을 생성하고, 추출된 조인트-스켈레톤 구조와 생성된 근육 및 외형 곡면을 이용하여 모델의 외형을 근사화한 표준 메쉬 모델을 생성하는 단계와, 동적 개체의 다시점(Multi-view) 영상 정보와 근사화된 표준 메쉬 모델의 투영(Projection) 영상 정보 사이의 오차가 허용 오차 이내가 되도록 파라미터를 제어하여, 표준 메쉬 모델의 전이를 수행하는 단계를 포함한다.
Abstract:
PURPOSE: An apparatus and a method for generating an animation are provided to generate a three-dimensional content animation by capturing the motion of an amorphous object. CONSTITUTION: A motion capturing unit(120) generates a mesh graph set about a recovery mesh set. The motion capturing unit generates motion generates motion data through the mesh graph set. The motion data includes motion information. A content generating unit(140) generates a mesh set through the motion data. The mesh set has a topology which is similar to the restoring mesh set.
Abstract:
PURPOSE: A motion capture apparatus and a method thereof are provided to generate three-dimensional frame by using previous frame information and to search for a feature point. CONSTITUTION: A three-dimensional appearance compensating unit(101) compensates three-dimensional shape by using a multi-view image. A three-dimensional framing unit(102) generates a three-dimensional frame by using an object model having a motion of a T frame. A motion compensating unit(103) compensates a motion of a T+1 frame by change of the motion.
Abstract:
PURPOSE: A dynamic object motion capture method and an apparatus thereof are provided to restore the motion information of a dynamic object motion using the image sensor of a visible light area. CONSTITUTION: An exterior information restoring unit(30) restores the geometric exterior information of a 3D model depending on the motion of a dynamic object. A skeleton approximation unit(40) approximates a 3D skeleton for the operation of a frame. A posture estimating unit(50) presumes the posture of the 3D model. According to difference between a multi-viewpoint composite image and a multi-viewpoint image, a shape estimating unit(60) presumes the shape of the 3D model.
Abstract:
본 발명은 동적 개체 모션 캡쳐 방법 및 그 장치에 관한 것으로, 골격계를 기반으로 골격계의 움직임을 통해 동작을 취하는 인체와 같은 동적 개체의 움직임 정보를 멀티카메라를 이용하여 촬영한 다시점 비디오 영상으로부터 동적 모션을 취하는 객체의 외형정보 및 주요 관절의 모션 정보를 복원하도록 한다. 본 발명에 따르면, 가시광 영역의 영상 센서만을 이용하여 동적 모션을 취하는 개체의 모션 정보를 복원할 수 있으며, 복원된 정보를 효율적으로 저장하여 다시점 영상을 재현할 수 있는 이점이 있다. 또한, 특정한 마커의 부착 없이 동적 개체의 모션 정보를 복원할 수 있는 이점이 있다.
Abstract:
PURPOSE: A real-time motion capturing device using parallel processing and a method thereof are provided to secure real-time features of motion reconstruction through parallel algorithm composition. CONSTITUTION: An elliptical model generation part(100) extracts the position of a joint from a volume model which is obtained from an initial posture multi-view image inputted from a pre-compensated multi-view camera, and generates a skeleton structure. The elliptical model generation part approximates the circumference of each human part to an elliptical model through elliptical fitting. A motion reconstruction part(200) reconstructs a motion from the multi-view image of each frame by using the generated elliptical model. [Reference numerals] (100) Elliptical model generation part; (200) Motion reconstruction part; (AA) Multi-view image obtained from initial posture; (BB) Elliptical model; (CC) Motion model
Abstract:
PURPOSE: An automatic skinning and appearance restoring device of a dynamic object and a method thereof are provided to automatically generate an intrinsic mesh model without geometric correction by using multiview image information including only appearance information of the dynamic object. CONSTITUTION: An image capturing unit(200) generates a multiview image of a dynamic object, multiview silhouette information, and a primary globally fitted standard mesh model based on a dynamic image object and a standard mesh model. A 3D image restoring unit(300) generates an NURBS-based intrinsic mesh model in the dynamic object about the primary globally fitted standard mesh. A data output unit(400) generates and outputs animation data and a final intrinsic mesh model. [Reference numerals] (100) Camera; (200) Image capturing unit; (300) 3D image restoring unit; (400) Skinning and skin data output unit; (AA) Standard mesh data; (BB) Final intrinsic mesh model; (CC) Animation data
Abstract:
PURPOSE: A shape transition method of a three-dimensional model and an apparatus thereof are provided to generate a three-dimensional model for expressing a motion of the object. CONSTITUTION: A multi-view camera unit obtains a plurality of two-dimensional images in a first frame(S200). A shape restoring unit of a shape transition controller generates a first volume model(S210). The shape transition controller generates a first shape transition model(S220). The multi-view camera unit obtains a plurality of two-dimensional images in a second frame(S230). The shape restoring unit generates a second volume model(S240).