Automated method for evaluating the attractiveness of an object, product program implementing such a method and computer support.

Abstract

The invention relates to a method for evaluating the attractiveness of an object comprising the following steps: A- defining a mesh (15) conforming to the shape of the object, comprising a plurality of nodes (vi), each node (vi) of the mesh (15) being associated with at least one geometric data point and at least one colorimetric data point; B - determining a geometric attractiveness value for each node (vi); C - determining a colorimetric attractiveness value for each node (vi); D- calculating an overall attractiveness value for each node (vi) based on the geometric attractiveness value and the colorimetric attractiveness value of said node (vi). Figure 1 for the abstract

Description

Title of the invention: Automated method for evaluating the attractiveness of an object, product program implementing such a method and computer support. technical field

[0001] The invention relates to the field of object design and the determination of the expected appeal of such an object within the framework of its design.

[0002] The invention thus relates to an automated method for evaluating the attractiveness of an object, a program product implementing such a method and a computer-readable computer storage medium storing such a program product. Prior state of the art

[0003] When designing an object, it is important to estimate how it will be perceived by consumers and to determine which elements of the object will be most attractive to them. Currently, this estimation is mainly based on the designer's experience and on tests with consumer focus groups. Since these tests must be carried out on each of the object's concepts, they can lead to relatively high costs and delay the product's entry into production.

[0004] To remedy this, it is possible to implement methods for determining visual perception such as that taught by Yang B. and his co-authors in their work published on May 16, 2015 in the journal "The Visual Computer," Volume 32, pages 1415 to 1432. Indeed, in this context, Yang B. and his co-authors demonstrate that it is possible to determine the areas of visual appeal of an object based on: - a mesh conforming to the shape of the object comprising a plurality of nodes, - a two-dimensional texture representing the color variations along the surface.

[0005] From each of these two elements, a first and a second map of the object's attention zones are determined individually and independently, based respectively on the object's geometry and colorimetry. The two maps thus determined can then be combined to provide a map of the object's visual attention zones that takes into account both the geometry, through the use of the mesh, and the colorimetry, through the use of the texture.

[0006] If such an automatic approach to visual perception could provide a While an automated validation solution for an object concept is not perfectly suited, it is because the consideration of geometric and colorimetric visual attention is completely independent of each other. Indeed, when a subject perceives an object, the influences of the object's geometry and colorimetry are inseparable, and it is important to be able to determine the object's visual appeal while taking this inseparability into account. Description of the invention

[0007] The invention thus aims to remedy the above drawback and has as its object a method for providing an evaluation of the attractiveness of an object taking into account both geometry and colorimetry without loss of information, in particular with regard to depth in three dimensions, as is the case with independent treatment of geometry and colorimetry as proposed by Yang B. and his co-authors.

[0008] The invention relates to this purpose to a method for evaluating the attractiveness of an object comprising the following steps: A- Definition of a mesh conforming to the shape of the object, comprising a plurality of nodes, each node of the mesh being associated with at least one geometric data point of said node and at least one colorimetric data point of a surface of the object at the level of said node, B- determination for each node of the mesh of a geometric attractiveness value calculated from the geometric data of said node (vz) and of first neighboring nodes of said node, C- determination, for each node of the mesh, of a colorimetric attractiveness value calculated from the colorimetry data of said node and its nearest neighbors. D- determination for each node of an overall attractiveness value from the geometric attractiveness value and the colorimetric attractiveness value of said node.

[0009] Thus, in the present method, the determination of geometric and colorimetric attractiveness is based on the same mesh. Consequently, there is no loss of information when combining geometric and colorimetric attractiveness to determine the overall attractiveness of each node. The overall attractiveness thus determined is more reliable and perfectly usable in the context of object design.

[0010] Step B of determining a geometric attractiveness value may include the following substeps: Bl- after having previously calculated for each mesh of the grid, and from the geometric data associated with each node vi of said mesh, a normal to said mesh, for each node determination of a normal from an average of the normals of the meshes to which said node belongs, B2- for each node and each of the nearest neighbor nodes, calculation of a difference value of normals between the normal determined in substep B1 for said nearest neighbor node and the normal determined in substep B1 for said node, B3- for each node, calculation of a difference value of normals from the difference values ​​of normals calculated in step B2 for said node and for its nearest neighbor nodes, said difference value of normals calculated for said node corresponding to a first intermediate geometric attractiveness value of said node.

[0011] Such sub-steps make it possible to determine a geometric attractiveness value based on the values ​​of the normals of the surface of the object at the level of each node of the mesh and thus to determine the areas of protrusion or indentation of the surface of the object which are likely to be attractive to the eye.

[0012] During substep Bl, the normal of each node can be represented by the components of an average normal vector calculated from the average of normal vectors of the meshes to which said node belongs, the substep of calculating a value of difference of normals calculated in substep B2 being a calculation of distance between the normal vector calculated for said first neighbor node in substep B1 and the average normal vector calculated for said node in substep Bl.

[0013] Such a determination based on a distance between the normals makes it possible to provide a calculation of the first value of geometric attractiveness which is particularly simple and reliable, since it is based on a calculation of the distance between the normal vectors.

[0014] Step B of determining a geometric attractiveness value may include the following substeps: B4. For each node and for each of the nearest neighbor nodes of said node, from the geometric data of said node and said nearest neighbor node, calculate a value of difference in geometric coordinates between said node and said nearest neighbor node, B5. For each node, calculate a geometric difference value from the coordinate difference values, said geometric difference value corresponding to a second intermediate geometric attractiveness value.

[0015] Such a geometric attractiveness value is also particularly reliable, since it is based on a geometric distance between the nodes. Indeed, a large distance corresponds to a large variation in the geometry of the surface and therefore to a high susceptibility to attracting attention.

[0016] Step B of determining a geometric attractiveness value may further include the following substep: B6- for each node, calculation of a geometric attractiveness value from the first intermediate geometric attractiveness value and the second intermediate geometric attractiveness value, said geometric attractiveness value being preferably equal to a weighted sum of the first intermediate geometric attractiveness value and the second intermediate geometric attractiveness value, the weighting factor used for this weighted sum being advantageously based on a roughness value previously calculated for said node.

[0017] Such a weighted sum makes it possible to obtain a particularly reliable geometric attractiveness value since it is based on both the first and second intermediate geometric attractiveness values. Furthermore, the weighting factor based on the roughness value can provide good consideration of the local variability of the object's surface topology.

[0018] Step C of determining a colorimetric attractiveness value for each node of the mesh, comprising the following sub-steps:

[0019] Cl- for each node and for each nearest neighbor node of said node (vz), calculation of a colorimetry difference between the colorimetry data of said node and that of said nearest neighbor node of said node,

[0020] C2- for each node, calculation of a colorimetric difference value from coordinate difference values, said colorimetric difference value corresponding to the colorimetric attractiveness value.

[0021] Such a determination based on a difference in colorimetry between the nodes makes it possible to provide a calculation of a colorimetric attractiveness value which is particularly simple and reliable.

[0022] For each node, at least one colorimetry data can include the components of a color associated with said node according to a color code, this color code preferably being the HSV system, for Hue, Saturation, Brightness.

[0023] Such use of a color code allows for a reliable calculation of the said color difference value. It should also be noted that when the HSV system is used, this value is particularly effective in providing a good estimate of color attractiveness.

[0024] The following step may also be envisaged: -E- Displaying a map of an attractiveness value from among the overall attractiveness value, the geometric attractiveness value and the colorimetric attractiveness value of the different nodes on a representation of the object, said attractiveness values ​​being preferably represented using a shade of grey scale or of color.

[0025] Such a display allows the user to perfectly identify the most attractive elements of the object to be evaluated.

[0026] The invention further relates to a computer program product comprising information code configured to implement the steps of a method for evaluating the attractiveness of an object according to the invention when executed on a computer.

[0027] Such a programmed product allows the implementation of a process according to the invention and thus to benefit from the advantages associated with it.

[0028] The invention also relates to a computer-readable computer storage medium storing the program product according to the invention. Brief description of the drawings

[0029] The present invention will be better understood upon reading the description of exemplary embodiments, given purely by way of illustration and in no way limiting, with reference to the accompanying drawings in which:

[0030] [Fig-1] illustrates a close-up example of the definition of a mesh conforming to the shape of an object within the framework of implementing a method for evaluating the attractiveness of an object according to the invention,

[0031] [Fig.2] illustrates a flowchart of the main steps of a process according to the invention,

[0032] [Fig.3] illustrates a flowchart of the substeps of a determination step B of a geometric attractiveness value for each node of the mesh,

[0033] [Fig.4] illustrates a flowchart of the substeps of a determination step C of a colorimetric attractiveness value for each node of the mesh,

[0034] [Fig.5] illustrates in perspective view an example of an object that can be evaluated within the framework of the process according to the invention.

[0035] [Fig.6A] illustrates the result of a determination of the geometric attractiveness values ​​obtained in the context of the implementation of a process according to the invention for the object illustrated in [Fig.5].

[0036] [Fig.6B] illustrates the result of a determination of the colorimetric attractiveness values ​​obtained in the context of the implementation of a process according to the invention for the object illustrated in [Fig.5].

[0037] [Fig.6C] illustrates the result of a determination of the overall attractiveness values ​​obtained in the context of the implementation of a process according to the invention for the object illustrated in [Fig.5].

[0038] Identical, similar or equivalent parts of the different figures bear the same numerical references so as to facilitate the transition from one figure to another.

[0039] The different parts shown in the figures are not necessarily according to a uniform scale, to make the figures more legible.

[0040] The different possibilities (variants and embodiments) should be understood as not being mutually exclusive and can be combined with each other.

[0041] Detailed description of particular embodiments

[0042] Fig. 1 illustrates a close-up view of a three-dimensional mesh used in the implementation of a method for evaluating the attractiveness of an object 10 according to the invention, both for determining a geometric attractiveness value and for determining a colorimetric attractiveness value.

[0043] Thus, a method for evaluating the attractiveness of an object 10 comprises, as illustrated by the flowchart in [Fig.2], the following steps: A- Definition of a mesh conforming to the shape of object 10 comprising a plurality of nodes vi, each node vi of the mesh being associated with at least one geometric data point of said node and with at least one colorimetry data point of the surface of object 10 at the level of said node, B- determination for each node vi of the mesh of a geometric attractiveness value calculated from the geometric data of said node vi and of first neighbor nodes vj of said node vi, C- determination for each node vi of the mesh of a colorimetric attractiveness value calculated from the colorimetry data of said node vi and of first neighbor nodes vj of said node, D- determination for each node vi of an overall attractiveness value from the geometric attractiveness value and the colorimetric attractiveness value of said node vi, E- display of maps of the overall, geometric and colorimetric attractiveness values ​​of the different nodes (vi) on a representation of object 10, the different attractiveness values ​​being represented using a shade of grey or colour scale, F- determination of geometric, colorimetric and global attractiveness scores from respectively the global, geometric and colorimetric attractiveness values.

[0044] In step A of defining the mesh, the mesh may be a linear triangular mesh, as illustrated in [Fig. 1]. Of course, such a mesh is provided only as an illustration, and other types of meshes, such as a quadrilateral mesh, whether linear or quadratic, may also be used. Similarly, while in the present embodiment the triangles have a substantially constant internode distance, it is also conceivable, without departing from the scope of the invention, that the mesh spacing be variable in order to take into account any significant variations in the surface of the object 10.

[0045] With a triangular mesh, as shown in [Fig. 1], each node vi has six nearest neighbor nodes vj and therefore six edges. In a practical implementation of the present method, to allow traversal of the different nodes vi of the mesh, it is possible to use an undirected graph. Thus, if we consider a mesh with V nodes, the undirected graph can be written as G = {V, E], where E is the number of edges of the graph.

[0046] Once the mesh of the surface of object 10 has been created, each node vi is associated with at least one geometric data and at least one colorimetry data.

[0047] At least one geometric datum may be geometric coordinates, such as Cartesian coordinates of said node associated with a reference frame attached to the object 10, or coordinates of another type, such as spherical coordinates, in association with a reference frame attached or not to the object 10 itself. Alternatively, such geometric datum

[0048] At least one colorimetric data point may be the values ​​associated with the surface color at said node, for example, the components of this color according to an RGB code (for Red, Green, and Blue, where each value corresponds to a percentage of the associated primary color) or the components according to the HSV code (for Hue, Saturation, Brightness). Of course, this color at the node may be a precise color determined on the surface of the object 10 at said node or an average color determined over an area of ​​the surface of the object 10 around said node, without departing from the scope of the invention.

[0049] More specifically, in step A, which provides a definition of a mesh conforming to the shape of object 10, a refined textured mesh can be implemented, that is, one offering a precisely adjusted association between the texture of the object's surface and the mesh. According to this approach, the association of colorimetric data for the surface of object 10 at each node vi can be achieved by remeshing each cell by subdividing them into sub-cells and sub-nodes, transferring the colors from the texture to the sub-nodes of this sub-cell, and then determining the colorimetric data based on the colors assigned to said sub-nodes. It should be noted that, according to this approach, the sub-cell replaces the mesh previously developed for the remaining steps of the process, the mesh used in steps B to F being the sub-cell obtained during the remeshing.

[0050] It should be noted that if, as indicated above, the colorimetric data are preferentially the components according to the HSV encoding (for Hue, Saturation, Brightness), the color data at the level of each node can be determined according to the RGB components and then converted according to the HSV encoding in order to be used as part of step C.

[0051] In the present embodiment, the geometric and colorimetric data used are respectively Cartesian coordinates and a code according to the TSV coding associated with said node (i.e. at the level of the node).

[0052] Figure 3 shows a flowchart of the main substeps of step B, which determines a geometric attractiveness value for each node vi of the mesh. It can thus be seen that step B can comprise the following substeps: Bl- after having previously determined, for each mesh element, and from the geometric data associated with each node vi of said mesh, a normal of said mesh, for each node vi, determination of a normal from an average of the normals of the mesh elements to which said node vi belongs, B2- for each node vi and each of the nearest neighbor nodes vj, calculation of a difference value of normals between the normal determined in sub-step B1 for said nearest neighbor node vj and the normal determined for said node vi in ​​sub-step B1, B3- for each node vi, calculation of a difference of normals value from the difference of normals values ​​calculated in step B2 for said node vi and for its nearest neighbor nodes vj, said difference of normals value calculated for said node vi corresponding to a first intermediate geometric attractiveness value of said node vi, B4. For each node vi and for each of the nearest neighbor nodes vj of said node vi, from the geometric data of said node vi and said nearest neighbor node, calculation of a value of difference of geometric coordinates between said node and said nearest neighbor node, B5. For each node vi, calculate a geometric difference value from the coordinate difference values, said geometric difference value corresponding to a second intermediate geometric attractiveness value, B6- for each node vi, calculation of a geometric attractiveness value from the first intermediate geometric attractiveness value and the second intermediate geometric attractiveness value multiplied by a roughness value determined for said node vi, said geometric attractiveness value being preferably equal to a weighted sum of the first intermediate geometric attractiveness value and the second intermediate geometric attractiveness value.

[0053] In the present embodiment, since the geometric data associated with the nodes are Cartesian coordinates, the normals of the cells previously determined in substep B1 can be normal vectors to the cells of the mesh, that is, perpendicular to the surface of these cells, along a direction directed away from the object 10 (direction external to the object). These normal vectors can be normalized to 1. Thus, according to this possibility, each normal for the cells is represented by three coordinates, each corresponding to an axis of the Cartesian coordinate system normalized to 1. Once the normals are determined for each cell, the normal for a given node vi can correspond

[0054] For each node vi and each of the nearest neighbor nodes vj, the difference in normals between the normal determined in substep B1 for said nearest neighbor node vj and that determined in the same step for node vi can be a distance between said corresponding normal vectors. Such a distance can be calculated from the square root of the sum of the squared differences of each of the coordinates of the corresponding normal vectors.

[0055] From these normal difference values, in substep B3, a normal difference value can be calculated for each node vi. This value can be an average of the normal difference values ​​calculated in step B2 for each of the nearest neighbor nodes vj of said node vi. Thus, such a normal difference value can be calculated by summing the normal differences obtained for the nearest neighbor nodes vj of node vi and dividing by the number of nearest neighbor nodes vj of said node vi. In other words, according to the distance calculation method mentioned in the paragraph above, the normal difference value for node vi can be an average of the distances calculated between node vi and each of its nearest neighbor nodes vj.

[0056] Such a value of difference of normals will be all the more important the more prominent or recessed the node vi is with respect to its nearest neighbor nodes vj. Therefore, this value of difference of normals is a good indicator of the visual attractiveness of said node vi and thus forms, within the framework of the present embodiment, a first intermediate geometric attractiveness value of said node vi.

[0057] Substep B4 allows, for each node vi and for each of the nearest neighbor nodes vj of said empty node, the calculation of a difference in geometric coordinates between said node vi and its nearest neighbor node vj. Since the geometric data associated with each node vi are Cartesian coordinates in this embodiment, such a difference in coordinates can be a calculation of the distance between said node and said nearest neighbor node. Such a distance can be calculated based on the square root of the sum of the squared differences of each of the coordinates of the node vi and its nearest neighbor node vj.

[0058] From these geometric difference values ​​calculated in substep B4, a geometric difference value can be calculated during substep B5 for each node vi. This geometric difference value can be an arithmetic mean of the geometric difference values ​​calculated between said node vi and its near-neighbor nodes vj. Similar to the difference in normals, a node vi exhibiting a relatively small geometric difference value corresponds to a relatively large protrusion or depression of the object 10, thus presenting visual appeal, whereas for a node located on a plane, this distance will be greater and have less visual appeal. Therefore, this geometric difference value forms, within the framework of the present embodiment, a second intermediate geometric attractiveness value for said node vi.

[0059] Once the first and second intermediate geometric attractiveness values ​​have been calculated, substep B6 calculates a geometric attractiveness value for each node vi based on these values. In this embodiment, the geometric attractiveness value can be a weighted sum of the first intermediate geometric attractiveness value and the second intermediate geometric attractiveness value multiplied by the weight of the first. In this embodiment, the weighting factor used for this weighted sum can be based on a roughness value determined for said node vi.

[0060] This roughness value determined for each of the nodes vi can be achieved for example on the basis of the teaching of K. Wang and his co-authors provided in their publication in the scientific journal "computer & Graphics" volume 36 number 7 pages 808 to 818 in November 2012.

[0061] To illustrate the result of such a geometric attractiveness value calculation, substeps B1 to B6 were implemented on object 10 shown in [Fig. 5], which represents a lion. This figure shows that the object has a texture with a light color gradient over most of the lion's body and a dark gradient on the mane, the tip of the tail, and the claws. After a preliminary meshing, in accordance with step A, this object was subjected to steps B1 and B6 to determine a geometric attractiveness value for each node of the mesh. The result of this submission is shown in [Fig. 6A] as a 3D map. This figure shows that the elements of object 10 with the greatest geometric attractiveness are the lion's mane, eyes, and tail tip.

[0062] Once the geometric attractiveness value has been determined in step B, step C, within the scope of the present invention, allows for the determination of a colorimetric attractiveness value. To do this, step C can implement, as shown in the flowchart illustrated in [Fig. 4], the following substeps: Cl- for each node vi and for each first neighbor node vj of said node vi, calculation of a colorimetry difference between the colorimetry data of said node and that of said first neighbor node vj of said node vi, C2 - For each node, calculate a colorimetric difference value from the coordinate difference values; said colorimetric difference value corresponding to the colorimetric attractiveness value

[0063] In the context of this embodiment, the colorimetric data of each node vi includes the components of the color associated with said node vi. The colorimetric difference calculated in substep Cl can be a calculation of the distance between node vi and one of its nearest neighbor nodes vj, or a calculation of the colorimetric distance between the color associated with said node and that associated with the nearest neighbor node vj. Such a distance calculation can be performed on the basis of a square root of the sum of the squared differences of each of the components of these colors.

[0064] The colorimetric differences calculated between each node vi and its nearest neighbor nodes vj can be used in substep C2 to calculate the colorimetric difference value, for example by taking, for each node vi, an arithmetic mean of the colorimetric differences calculated in substep C1 for all the nearest neighbor nodes vj of said node vi. This colorimetric difference value constitutes the colorimetric attractiveness value.

[0065] In a manner identical to the geometric attractiveness calculation, to illustrate the result of such a colorimetric attractiveness value calculation, substeps C1 to C2 were implemented on object 10 in [Fig. 5]. After a preliminary meshing, in accordance with step A, this object was subjected to steps C1 and C2 to determine a colorimetric attractiveness value for each node of the mesh. The result of this submission is illustrated in [Fig. 6B] as a 3D map. It can be seen in this figure that the elements of object 10 exhibiting the greatest geometric attractiveness are also the lion's mane, eyes, and tail tip, in addition to the lion's claws.

[0066] Once the geometric and colorimetric attractiveness values ​​have been determined, they can be combined in step D to determine an overall attractiveness value. In this embodiment, this determination can be made for each node as a maximum value between the geometric and colorimetric attractiveness values, so as to identify each attractiveness zone of the object 10, whether this attractiveness is related to the geometry of the object 10 or to its colorimetry.

[0067] Of course, the use of such a maximum value is given only by way of example, and other combinations of geometric and colorimetric attractiveness values ​​to determine an overall attractiveness value are conceivable without departing from the scope of the invention. Thus, for example, if it is desired to favor areas exhibiting both a high geometric and colorimetric attractiveness value, the attractiveness value can be calculated for each node 10 based on the square root of the multiple of the geometric attractiveness value and the colorimetric attractiveness value.

[0068] In order to illustrate the calculation of such an overall attractiveness value, based on the object 10 shown in [Fig.5] and the geometric and colorimetric attractiveness values ​​shown in Figures 6A and 6B, [Fig.6C] illustrates, in the form of a 3D map, the overall attractiveness values ​​33 determined in the context of such an implementation of step D. In accordance with what had been identified with the geometric and colorimetric attractiveness values, the elements of the object 10 which exhibit the greatest overall attractiveness are the mane, the eyes, the claws and the tip of the lion's tail.

[0069] Once the geometric, colorimetric and global attractiveness values ​​have been determined for each of the nodes vi, it is possible, in the context of steps E and F, to display the result in the form of 3D maps, as shown in figures A to C, and to determine the associated attractiveness scores.

[0070] It should be noted that in step F, the geometric, colorimetric, and global attractiveness scores can be calculated by taking an arithmetic mean over all the nodes of the respective geometric, colorimetric, and global attractiveness values. Of course, if this score provides a good estimate of the visual attractiveness of object 10, it is possible to provide other values ​​of interest based on the attractiveness values ​​determined during steps B, C, and D, such as, for example, an attractiveness variance or a median value.

[0071] By way of illustration of such scores, in the present example of the object shown in [Fig. 5], the scores obtained are respectively 0.31 for geometric attractiveness, 0.50 for colorimetric attractiveness, and 0.54 for overall attractiveness. Thus, while there is a synergy between geometric and colorimetric attractiveness, it is colorimetric attractiveness that is primarily responsible for the overall attractiveness score obtained for this object 10.

[0072] It should be noted that although the main object of the present invention is a method for evaluating the attractiveness of an object 10, this can of course be implemented in the context of the manufacture of an object 10. According to this possibility, the invention also relates to a method for manufacturing an object 10 comprising the following steps: - provision of a plurality of object prototypes, - determination for each of the object prototypes and from the attractiveness evaluation method of the attractiveness values ​​and scores, - selection of the object 10 to be manufactured from among the object prototypes from the overall attractiveness values ​​and scores determined during the implementation of the attractiveness evaluation method according to the invention, - manufacture of the chosen object 10.

[0073] This manufacturing step of the object 10 can in particular be carried out by additive manufacturing in order to allow the object to be produced in a relatively short period of time reduced. In this way, the development of a new object, from design to the manufacture of a first prototype of the object validated by the attractiveness evaluation process according to the invention, can take less than a week, whereas these steps require several months in the context of a classic manufacturing process.

Claims

Claims

1. A method of evaluating the attractiveness of an object (10) comprising the following steps: A. definition of a mesh (15) matching a shape of the object (10) comprising a plurality of nodes (vi), each node (vz) of the mesh (15) being associated with at least one geometric data item of said node and at least one colorimetry data item of a surface of the object (10) at said node, B. determination for each node (vz) of the mesh (15) of a geometric attractiveness value calculated from the geometric data of said node (vz) and first neighbor nodes (vj) of said node (vi), C. determination for each node (vz) of the mesh (15) of a colorimetric attractiveness value calculated from the colorimetry data of said node (vz) and the first neighbor nodes (vj) of said node,, D. determination for each node (vz) of a global attractiveness value from the geometric attractiveness value and the colorimetric attractiveness value of said node (vz).

2. A method of evaluating the attractiveness of an object according to claim 1, wherein step B of determining a geometric attractiveness value comprises the following substeps: Bl- after having previously calculated for each cell of the mesh (15), and from the geometric data associated with each node vi of said cell, a normal to said cell, for each node (vz) determination of a normal from an average of the normals of the cells to which said node (vz) belongs, B2- for each node (vz) and each of the first neighbor nodes (vj), calculation of a normal difference value between the normal determined in sub-step B1 for said first neighbor node (vj) and the normal determined in sub-step Bl for said node (vi), B3- for each node (vi), calculation of a normal difference value from the normal difference values ​​calculated in step B2 for said node (vi) and for its first neighbor nodes (vj ), said normal difference value calculated for said node (vi) corresponding to a first intermediate geometric attractiveness value of said node (vi).

3. Method for evaluating the attractiveness of an object according to claim 2, wherein during the sub-step of B1, the normal of each node (vi) is represented by the components of an average normal vector calculated from the average of normal vectors of the meshes to which said node (vi) belongs, and wherein, the sub-step of calculating a normal difference value calculated in sub-step B2 is a calculation of distance between the normal vector calculated for said first neighbor node in sub-step B1 (vj) and the average normal vector calculated for said node (vi) in sub-step Bl.

4. A method for evaluating the attractiveness of an object according to any one of claims 1 to 3, wherein step B of determining a geometric attractiveness value comprises the following sub-steps: B4. for each node (vi) and for each of the first neighbor nodes (vj) of said node (vi), from the geometric data of said node (vi) and said first neighbor node, calculating a geometric coordinate difference value between said node and said first neighbor node, B5. for each node, calculating a geometric difference value from the coordinate difference values, said geometric difference value corresponding to a second intermediate geometric attractiveness value.

5. Method for evaluating the attractiveness of an object according to claims 2 and 4 alone or in combination with claim 3, wherein step B of determining a geometric attractiveness value further comprises the following sub-step: B6- for each node (vi), calculating a geometric attractiveness value from the first intermediate geometric attractiveness value and the second intermediate geometric attractiveness value, said geometric attractiveness value preferably being equal to a weighted sum of the first intermediate geometric attractiveness value and the second intermediate geometric attractiveness value, the weighting factor used for this weighted sum being advantageously based on a roughness value previously calculated for said node (vi).

6. A method of evaluating the attractiveness of an object according to any one of claims 1 to 5, wherein step C of determining for each node (vi) of the mesh (15) of a colorimetric attractiveness value comprising the following sub-steps: Cl- for each node (vi) and for each first neighbor node (vj) of said node (yï), calculation of a colorimetry difference between the colorimetry data of said node and that of said first neighbor node (vj) of said node (vz), C2- for each node (vz), calculation of a colorimetry difference value from the coordinate difference values, said colorimetry difference value corresponding to the colorimetric attractiveness value.

7. Method for evaluating the attractiveness of an object according to claim 6, in which for each node (vi), the at least one colorimetry data item comprises the components of a color associated with said node according to a color code, this color code preferably being the TSV system, for Hue, Saturation, Brightness.

8. Method for evaluating the attractiveness of an object according to any one of claims 1 to 7, in which the following step is further provided: A. Displaying a mapping of an attractiveness value among the overall attractiveness value, the geometric attractiveness value and the colorimetric attractiveness value of the different nodes (vi) on a representation of the object, said attractiveness values ​​being preferentially represented using a gray or color shade scale.

9. A computer program product comprising information code configured to implement the steps of a method for evaluating the attractiveness of an object according to any one of claims 1 to 8 when executed on a computer.

10. A computer-readable computer recording medium storing the program product according to claim 9.

Images