Abstract: A method of reinforcement learning of a neural network device for generating a verification vector for verifying a circuit design comprising a circuit block includes inputting a test vector to the circuit block, generating one or more rewards based on a coverage corresponding to the test vector, the coverage being determined based on a state transition of the circuit block based on the test vector, and applying the one or more rewards to a reinforcement learning.

Abstract: Disclosed herein is a motion modeling method and device using a virtual joint sphere. The motion modeling method using the virtual joint sphere includes (a) modeling a 3D virtual character, wherein the 3D virtual character sets a virtual joint sphere for each body part; and (b) forming a trajectory for a motion of a target body part of the 3D virtual character as a specific motion is performed on a surface of the virtual joint sphere assigned to the target body part, when reconstructing the motion of the 3D virtual character according to the specific motion.

Abstract: Disclosed herein is a motion modeling method and device using a virtual joint sphere. The motion modeling method using the virtual joint sphere includes (a) modeling a 3D virtual character, wherein the 3D virtual character sets a virtual joint sphere for each body part; and (b) forming a trajectory for a motion of a target body part of the 3D virtual character as a specific motion is performed on a surface of the virtual joint sphere assigned to the target body part, when reconstructing the motion of the 3D virtual character according to the specific motion.

Abstract: Three-dimensional modeling apparatus and method, the three-dimensional modeling apparatus including: an input curved surface; an input point arrangement unit; a representative point determining unit; a polygon generating unit; a curved surface model generating unit; and a curved surface model combining unit. A plurality of curved surface models is combined with one another to generate a three-dimensional image so that a three-dimensional shape that is similar to an actual shape of an object to be modeled may be easily generated.

Abstract: Three-dimensional modeling apparatus and method, the three-dimensional modeling apparatus including: an input curved surface generating unit generating a plurality of input curved surfaces each of those is formed of a plurality of input points based on coordinates and a moving direction of an input device that moves in three-dimensional data input space; an input point arrangement unit generating three-dimensional coordinate space that corresponds to the data input space and is formed of a plurality of unit grids, and arranging the input points in each of the unit grids; a representative point determining unit determining a representative point of each of the unit grids based on coordinates of the input points included in each of the unit grids and generating a group of representative points with respect to each of the input curved surfaces; a polygon generating unit selecting a representative point sequentially from representative points as a central representative point and generating polygon sequentially w

Abstract: Disclosed herein is an apparatus and method for deforming a curved surface using a three-dimensional target curve. The curved surface deformation apparatus of the present invention includes object curved surface selection means, sample vertex extraction means, basis function calculation means, target curve input means, control point variation calculation means, and object curved surface deformation means. The object curved surface selection means selects all or part of the curved surface of the NURBS model. The sample vertex extraction means extracts a plurality of vertexes from the object curved surface. The basis function calculation means calculates a basis function of the object curved surface. The target curve input means inputs a target curve, composed of a plurality of target. The target curve projection means projects the target curve onto the object curved surface.

Abstract: Disclosed herein is an apparatus and method for deforming a curved surface using a three-dimensional target curve. The curved surface deformation apparatus of the present invention includes object curved surface selection means, sample vertex extraction means, basis function calculation means, target curve input means, control point variation calculation means, and object curved surface deformation means. The object curved surface selection means selects all or part of the curved surface of the NURBS model. The sample vertex extraction means extracts a plurality of vertexes from the object curved surface. The basis function calculation means calculates a basis function of the object curved surface. The target curve input means inputs a target curve, composed of a plurality of target. The target curve projection means projects the target curve onto the object curved surface.