JPH0785313A - Deformed shape generation device - Google Patents

Abstract

PURPOSE:To facilitate the processing of shape deformation by performing matching to two shape models from a common basic shape, obtaining the feature amounts of the two shape models and interpolating the feature amount of both models at the time of deforming one model into the other one. CONSTITUTION:Shape data A301 and shape data B302 generate the feature amount A305 and the feature amount B306 from basic shape data 304 by a feature amount analysis unit 303. The feature amounts A305 and B306 are made to correspond to each other by using basic shape data 304 by a feature amount corresponding unit 307 and an interpolation amount 308 is given. Thus, deformed shape data 310 consisting of the features of both models is generated by a feature amount synthesis unit 309. Thus, a calculation value can be defined by a process based on the continuous change of the feature amounts even in the conversion of the shapes, and the expression restriction of the shape model in deformed shape generation can be avoided.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a shape deformation or deformed shape image generation apparatus for defining a new shape model by continuous deformation between two shapes.

[0002]

2. Description of the Related Art Conventionally, as this kind of shape deformation method,
The vertex list and the linked list for forming a face (line) between the two shapes should have exactly the same structure, and the structure should not be changed during continuous deformation between the two shapes. Interpolation was performed between the vertices of two shapes.

For this reason, when transforming a simple shape into a complicated shape, even for a simple shape, a vertex list that is exactly the same as that for expressing a complicated shape and a linked list regarding vertices are prepared. There is a drawback that it is necessary and the shape representation constraint is large (eg Terzopoulo
s, D., and K. Fleischer, “Modeling Inelastic Defromat
ion: Viscoelasticity, Plasticity, Fracture, ”SIGGRAP
See H'88, 269-278).

[0004]

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned conventional problems, and an object of the present invention is to generate an intermediate shape model by continuous deformation between shape models. It is an object of the present invention to provide a technique capable of facilitating the process of shape deformation in the deformed shape generation device.

The above and other objects and novel constitution of the present invention will become apparent from the description of this specification and the accompanying drawings.

[0006]

In order to achieve the above object, according to the present invention, in order to represent the shape of an object, two vertex lists and a linked list that connects the vertices to form a surface or line are provided. In a deformed shape generation device that generates an intermediate shape model by continuous deformation between shape models, with respect to both expressions of the shape model, a deformation from a common basic shape is performed once, the execution result is recorded, and From the first means for obtaining the feature amount of the shape model, the second means for interpolating between the one feature model and the other feature model when transforming the other feature model, and the interpolated feature amount And a third means for creating a concatenation list that forms a surface or a line by connecting the vertices of the intermediate shape model that represents the deformed shape. .

Further, in the above-mentioned means, a vertex list defining a shape and shape energy at each vertex are used as the feature quantity of the shape model.

[0008]

According to the above-mentioned means, the two basic shape models are matched with each other from the common basic shape, and the feature quantity for expressing the two geometric models is obtained.
When transforming to the other shape model, interpolation is performed on both feature amounts, so it is necessary to hold a shape model directly connected to both shapes and perform interpolation based on that shape model as in the conventional case. Absent.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a block diagram showing a schematic configuration of a deformed shape generating apparatus according to an embodiment of the present invention.

In FIG. 1, 301 is shape data A, 3
02 is shape data B, 304 is basic shape data, and 310
Is a deformed shape data, 305 is a feature amount A, 306 is a feature amount B, 303 is a feature amount analysis unit, 307 is a feature amount corresponding unit, and 309 is a feature amount combining unit.

The deformed shape generating apparatus of this embodiment shown in FIG. 1 is composed of, for example, an information processing apparatus such as a computer.

Shape data A301 and shape data B3
02, the characteristic amount analysis unit 303 generates a characteristic amount A 305 and a characteristic amount B 306 from the basic shape data 304, respectively.

These characteristic amount A305 and characteristic amount B3
In the case of 06, the feature quantity correspondence unit 307 takes correspondence using the basic shape data 304 and gives the interpolation quantity 308, and the feature quantity synthesizing unit 309 generates the deformed shape data 310 consisting of both features.

Next, the operation of the feature quantity analysis unit 303 will be described with reference to FIGS.

In the following description, an example in which a vertex list and a set of shape energy parameters at vertices are used as feature amounts in the case of a three-dimensional shape will be described.

For a simple vertex list and the shape energy parameters at the vertices, a shape model having three vertices around one vertex will be used as the shape model expressing the feature amount.

Further, as an example of the modification for obtaining the feature amount, a case where a spherical shape is used as the basic shape data will be shown.

FIG. 2 shows a shape model in which three vertices exist around one vertex, and with respect to the point Pi, the angle between the adjacent vertex and the adjacent surface is used as an energy parameter.
It is a figure which shows that the feature-value can be expressed.

As shown in FIG. 2, in a shape model in which three vertices (P _{N1 (i)} , P _{N2 (i)} , P _{N3 (i)} ) exist around one vertex (Pi), with respect to the point Pi, The feature amount can be expressed by the following equation using the angle between the adjacent vertex and the adjacent surface as an energy parameter.

[0021]

## EQU1 ## Pi = F + L (ri, di, φi) Ni FIGS. 3 and 4 are diagrams showing modifications of the vertex list expressing the feature amount.

The vertex list expressing the features is shown in FIGS.
As shown in FIG.

In FIG. 3, 100 is two small faces, 1
01 is a slightly large surface, and the two small surfaces 100 are converted into a slightly large surface 101 by the edge deletion 104, and the slightly large surface 101 is converted into a small surface 2 by the surface division 105. Is converted to 100.

In FIG. 4, 102 is a small surface 2 which is separated.
103 is a large surface, and the separated surface 102 is converted into a large surface 103 by the structure synthesis 106, and the large surface 103 is divided into two separated small surfaces 102 by the structure division 107. .

Next, referring to FIGS. 5, 6 and 7, a method of obtaining the feature amount from the basic shape will be described.

FIGS. 5, 6 and 7 are diagrams for explaining how to obtain the feature amount from the basic shape in this embodiment.

In FIG. 5, reference numeral 200 is a human head shape which is shape data, and 201 is a spherical shape which is basic shape data.

In FIG. 6, reference numeral 204 denotes a target shape surface, and 20
5 is a basic shape surface, and 206 is energy.

In FIG. 7, 207 is a matching surface, 208 is a feature vertex, and 209 is a shape energy parameter.

The spherical shape 201, which is the basic shape data shown in FIG. 5, is matched with the human head shape 200, which is the shape data, to obtain the feature shape 202, and the sides shown in FIGS. 3 and 4 are deleted. The deformed shape 203 can be obtained through 104, plane division 105, structure synthesis 106, and structure division 107.

Here, the spherical shape 2 which is the basic shape data
01, the surface of the person's head shape 200 is matched with the shape 202 of the feature, and the target shape surface 20 shown in FIG.
4, the spherical surface 201 which is the basic shape data.
Is the basic shape surface 205 shown in FIG. 6, the difference between the two is the energy amount 206.

When the surface of the shape 203 obtained by the conversion is used as the matching surface 207 shown in FIG. 7 and the vertices forming the surface are used as the characteristic vertices 208, the shape energy parameter 209 is thus represented by the vertex list. It is the intersection angle of the faces created by adjacent vertices at each vertex when the shape energy is minimized.

As described above, the shape data A301 is
In the case of the human head shape 200, the spherical shape 201 which is the basic shape data is replaced with the human head shape 20 which is the shape data.
When matching 0, the spherical shape 20 which is the basic shape data
1 and the human head shape 200 do not match, the energy amount 206 is large. However, by matching the energy 206 so that it is in the minimum state, the shape 202 via the feature corresponding to the human head shape is passed. And feature 2 by conversion
03 is obtained, which is the set of vertex list and shape energy parameters.

Similarly, for shape B, a vertex list and a set of shape energy parameters are created.

Then, these become the characteristic amount A305 and the characteristic amount B306.

FIG. 8 shows a characteristic amount A305 and a characteristic amount B3.
It is a figure which shows the table of 06.

In FIG. 8, the characteristic amount A305 and the characteristic amount B306 are expressed based on the vertex list of the basic shape data.

Next, the operation of the feature amount correspondence unit 307 and the feature amount combining unit 309 will be described.

From the shape data A301 to the shape data B3
In order to transform it into 02, the vertex list and shape energy parameter of the feature quantity A305 expressing the shape data A301 are converted into the feature quantity B30 expressing the shape data B302.
Consider moving to a vertex list of 6 and shape energy parameters.

In this embodiment, since the shape energy parameter 209 is the intersection angle of the faces of the adjacent vertices, the number of elements of the shape energy parameter in the characteristic amount A305 and the number of elements of the shape energy parameter in the characteristic amount B306 are equal to each other. Are different, and the shape data A30
Interpolation cannot be performed with the elements of the vertex list of the feature amount A305 expressing 1 and the set of shape energy parameters.

Therefore, the feature quantity correspondence unit 307 uses the vertex list of the basic shape data 304 so that the elements of the feature quantity A305 and the feature quantity B306 can be accessed to represent the shape data A301. Interpolation can be performed without changing the elements of the vertex list of the feature amount A305 and the set of shape energy parameters.

In the feature quantity synthesis unit 309, the interpolation quantity 3
Interpolation in the feature amount is performed according to the ratio of 08.

That is, the interpolation amount is α (0 ≦ α ≦ 1),
A new characteristic amount C is obtained by the following mathematical formula.

[0044]

## EQU00002 ## Feature amount C = .alpha. Feature amount A + (1-.alpha.) Feature amount B FIG. 9 shows a table of the feature amount C.

In the present embodiment, as described above, since the shape energy parameter 209 is the intersection angle of the faces of the adjacent vertices, the number of elements of the shape energy parameter in the characteristic amount A305 and the shape energy in the characteristic amount B306 are large. It differs from the number of elements in the parameter,
This interpolation alone is not enough.

Therefore, in the feature quantity synthesizing unit 309, the feature quantity A30 is obtained by the conversion shown in FIGS.
The number of elements expressing 5 and the number of elements expressing the feature amount B306 are interpolated.

As a result, a set of feature points (vertex list and shape energy parameter) having an appropriate number of elements can be generated.

By performing an operation reverse to that of the feature quantity analysis unit 303 from the feature points thus generated as a set, the deformed shape data to be obtained is generated.

Note that features other than the vertex list and shape energy parameters may be used as the feature amount.

Although the present invention has been specifically described based on the embodiments, it is needless to say that the present invention is not limited to the embodiments and various modifications can be made without departing from the scope of the invention.

[0051]

As described above, according to the present invention,
When the two basic shape models are matched with each other from a common basic shape, a feature quantity for expressing the two geometric models is obtained, and when the one geometric model is transformed into the other geometric model, the features of the two geometric models are compared. Since the interpolation is performed based on the quantity, it is not necessary to hold the shape model directly connected to the shapes of both and perform the interpolation based on the shape model as in the conventional case.

Therefore, even in the conversion between shapes,
The calculation amount can be defined by a process based on the continuous change of the feature amount, and it is possible to avoid the representation constraint of the shape model in the deformation shape generation.

[Brief description of drawings]

FIG. 1 is a block diagram showing a deformed shape generation device that is an embodiment of the present invention.

FIG. 2 is a diagram showing that, in a shape model in which three vertices exist around one vertex, the feature amount can be expressed with respect to a point Pi by using an angle between an adjacent vertex and an adjacent surface as an energy parameter.

FIG. 3 is a diagram showing a modification of a vertex list expressing a feature amount.

FIG. 4 is a diagram showing a modification of a vertex list expressing a feature amount.

FIG. 5 is a diagram for explaining how to obtain a feature amount from a basic shape in the present embodiment.

FIG. 6 is a diagram for explaining how to obtain a feature amount from a basic shape in the present embodiment.

FIG. 7 is a diagram for explaining how to obtain a feature amount from a basic shape in the present embodiment.

FIG. 8 is a diagram showing a table of a feature amount A305 and a feature amount B306 in the present embodiment.

FIG. 9 is a diagram showing a table of feature amounts C in the present embodiment.

[Explanation of symbols]

100 ... two small faces, 101 ... slightly large face, 102
... two small faces that are apart from each other, 103 ... large face, 104 ... edge deletion, 105 ... face division, 106 ... structural synthesis, 107
... structure division, 200 ... human head shape, 201 ... basic shape, 202 ... feature based shape, 203 ... transformed shape, 204 ... target shape surface, 205 ... basic shape surface,
206 ... Energy amount, 208 ... Feature vertex, 301 ... Shape data A, 302 ... Shape data B, 303 ... Feature amount analysis unit, 304 ... Basic shape data, 305 ... Feature amount A, 306 ... Feature amount B, 307 ... Feature Quantity correspondence unit,
308 ... Interpolation amount, 309 ... Feature amount combining unit, 310
… Deformed shape data.

Claims (2)

[Claims] 1. A transformation for generating an intermediate shape model by continuous transformation between two shape models having a vertex list and a linked list forming a surface or a line connecting the vertices to represent the shape of an object. In the shape generator,
With respect to both representations of the shape model, a deformation is performed once from a common basic shape, the execution result is recorded, and a first means for obtaining the feature amount of each shape model and one shape model from the other When transforming into a shape model, a second means for interpolating the two feature quantities, a vertex list of the intermediate shape model expressing a deformed shape and a surface connecting the vertices from the interpolated feature quantity Or a third means for creating a linked list forming a line, the deformed shape generating apparatus. 2. The deformed shape generating apparatus according to claim 1, wherein a vertex list defining a shape and shape energy at each vertex are used as the feature quantity of the shape model.