Sony Corporation

Junji Horikawa

Takashi Totsuka

Polygonal data input in a first step is subjected to evaluation in which all edges of the polygon data are ranked in importance on the basis of a volume change caused by removal of that edge. The edges are sorted on the basis of an evaluation value in a third step. In a fourth step, the edge of a small evaluation value is determined to be an edge of a small influence on the general shape and is removed. In a fifth step, a new vertex is determined from the loss of vertex by the edge removal. In a sixth step, a movement of texture coordinates and a removal of the texture after the edge removal are executed on the basis of the area change of the texture due to the edge removal by a predetermined evaluating function. In a seventh step, by repeating the processes in the second to sixth steps, a polygon model approximated to a desired layer can be obtained.

Computer animation generator

An shape data approximation method for reducing the amount of data with an adjacency relationship between a plurality of components maintained. The shape data is read, an adjacent region of the shape data is designated, a virtual surface and a virtual edge are created in the designated adjacent region, and each edge in the shape data is evaluated to determine a evaluation score. The evaluation score is based on a change in the shape of the adjacent region on the assumption that the virtual edge is removed. The edges are subjected to data sorting, and an edge having a low evaluation score is removed with high priority, and resulting approximate shape data is thus output.

Approximation method of shape data, information processing apparatus and medium

Distortion of normal vectors when approximating shape data with normal vectors is restrained. At step ST,original polygon date is input. At step ST,each edge is evaluated for performing removal of an edge. At step ST,the evaluation values are sorted by magnitude and the edge having the minimum value is selected. At step ST,the edge having the minimum evaluation value selected at step ST,is removed. At step ST,the position of vertices remaining after removal of the edge is decided. At step ST,a normal vector which has become unnecessary on removal of the edge is removed and the remaining normal vectors are changed. At step ST,an approximated model is drawn. The processing from steps ST,to ST,is repeatedly executed in accordance with a desired precision.

Image data approximation considering normal vectors

Distortion of normal vectors when approximating shape data with normal vectors is restrained. At step ST1, original polygon date is input. At step ST2, each edge is evaluated for performing removal of an edge. At step ST3, the evaluation values are sorted by magnitude and the edge having the minimum value is selected. At step ST4, the edge having the minimum evaluation value selected at step ST3 is removed. At step ST5, the position of vertices remaining after removal of the edge is decided. At step ST6, a normal vector which has become unnecessary on removal of the edge is removed and the remaining normal vectors are changed. At step ST7, an approximated model is drawn. The processing from steps ST2 to ST6 is repeatedly executed in accordance with a desired precision.

A method and apparatus for approximating shape data in which the degree of fineness of a desired portion is set so as to differ from that of other portions for hierarchical approximation by a facilitated designating operation. Polygon data of an original are entered and the partial degree of fineness of the polygon model is designated by the user. It is designated to which degree the model is approximated and each edge is evaluated for removing edges. The evaluated values are sorted in accordance with the magnitudes of the evaluated values and controlled responsive to designation by the user of the partial degree of approximation. An edge with the least evaluated value is selected and removed and the position of the apex points left after removing the edge are determined to see whether or not approximation has reached a desired degree of approximation. An approximated model is selected from plural hierarchies and processed for drawing or saving.

Method for approximating shape data, drawing apparatus and information

A drawing method of computer graphics in which an approximation of a model is executed in consideration of a target point of the observer, a position of model, a size of model on a picture plane, and a moving speed of model are considered and, further, a feeling of physical disorder that is given to the eyes of a human being is reduced is provided. Geometric model data inputted in step S1 is subjected to a filtering process in order to reduce a complexity of the model in step S2 and an outline of the model is extracted in step S3. As for the model from which the outline was extracted, characteristic points to which the human being largely react are extracted from such an outline in step S4. The processes in steps S2 to S4 are executed by a number of times as many as the number of necessary hierarchies (step S5). The characteristic points extracted every hierarchy in step S4 are made correspond among the hierarchies in step S6. The geometric model in a proper hierarchy is selected from the hierarchical approximated models obtained by the procedure in steps S2 to S6 in accordance with a state of the input geometric model in step S7. The selected geometric model is drawn as a computer graphics picture plane in step S8.

Growing community of inventors

Tokyo, Japan

Junji Horikawa