Even though the increased computational power have enabled users to observe heavy 3D polygonal animations on their desktops, there are still few effective methods for interactive online control of 3D models in cyberspace. In this paper, we propose a new inverse kinematics method by which users can control tree-structured 3D polygonal models with motion data. The key idea is to extract parameters necessary for inverse kinematics control from the motion data: such data include the mass matrix that determines the motion of the redundant joints, and the transform function of end effectors that determine the relative velocity of the controlled segment and the end effectors. Then such parameters are used to calculate the response of the 3D character model when users click and drag the parts of the model with their mouse. Using our method, users can easily edit, warp, and retarget 3D character motions by least effort. Our method is especially effective to handle complex 3D human motion data captured by motion-capture devices.

View full abstract

An efficient algorithm for searching similar images is proposed. The algorithm is applicable to content based image retrieval and multiple template matching. For designing the algorithm, the upper land lower limits of the correlation coefficients between registered images and a object image by an indirect way through a pibot image. An interval estimation based on the limits shows the range of an unknown but estimated correlation coefficient. Using the interval estimatin, some conditions for selecting redundant images of smaller similarity from the registered images can be detected and skipped away from conmputation, which enable to effeciently save computational cost of searching. The algorithm is applied to real searching problems in a image database of 1,200 images and in multiple template matching problems, showing the effectiveness of the proposed method.

View full abstract

We present an approximate object modeling method which makes use of “basic deformations” understandable for non-expert users. By using the basic deformations as the shape features of modeling, the shapes and deformations of objects are represented simultaneously. This helps users to recognize current shapes, and anticipate deforming results. In this study, a rectangular prism was chosen as the original shape. An extended generalized cylindrical model (GCM) is used to represent a shape, and two kinds of deformations are treated: (1) an axis-based deformation that can be described by the changes of GCM parameters; and (2) a face-based deformation the direction of which is determined according to the principal axis. Among these kinds of deformations, five basic deformations are contrived experimentally from a questionnaire survey result. Finally, we construct a modeling system using ten operations to attain the basic deformations and show our method's feasibility for an approximate modeling.

View full abstract