Method and device for modeling a real object

Abstract

The invention relates to a method for modeling a real object, comprising the steps of: Obtaining a two-dimensional and / or three-dimensional dataset related to the object, the dataset being associated with a first space; Generating a three-dimensional representation of the object in a second space from the dataset; The three-dimensional representation comprises a plurality of parts whose surfaces are generated by a mathematical function describing a parametric surface. The step of generating the three-dimensional representation of the object in the second, or canonical, space includes a sub-step of adjusting the shape and dimensions of each part to at least one element of the real object, the adjustment being made taking into account at least one structural constraint of said object. (See Figure 3 for the abbreviated version.)

Description

Title of the invention: Method and device for modeling a real object. TECHNICAL FIELD OF THE INVENTION

[0001] The field of the invention is that of the modeling of an object.

[0002] More specifically, the invention relates to a method and device for modelling a real object.

[0003] The invention finds particular applications in augmented reality to improve the rendering of a three-dimensional representation of the real object, for example in the context of a virtual fitting of the object by an individual. STATE OF THE ART

[0004] It is known in the prior art to model an object in a so-called modeling space that provides a three-dimensional representation of the object. These techniques are typically based on image processing and / or scanning the object by varying the viewing angle. Generally, the object representation obtained by these techniques is a single, discrete model without any segmentation between the different elements that make up the object. For example, in the case of a pair of glasses, the three-dimensional representation obtained from a scan of the glasses mounted on a frame does not differentiate the various elements such as the temples, the temples, the rims, the lenses, the bridge, or even the frame itself. All these elements are delimited by a point cloud forming a single three-dimensional surface without any differentiation between them.This also occurs when scanning a pair of shoes where the laces are fused with the upper and / or the tongue of each shoe. Similar difficulties are encountered when scanning a handbag with handles, a strap, or a chain that tend to fall down and become fused with the outer lining of the bag, without distinguishing between the two parts.

[0005] Discretizing the points in the cloud leads, on the one hand, to a cumbersome file size and, on the other hand, to inaccuracies in the modeled surfaces, or even to lighting artifacts in the rendering, particularly when dealing with a translucent texture. Biases can also be introduced, such as smoothing of details or the introduction of artifacts. In the example of the scanned pair of glasses, the smoothing of details is particularly visible at the tenons, where the different parts of the tenon, such as the screws or the hinge, are merged together and therefore unidentifiable. Artifacts are also introduced at the point of contact between the support and the temple of the glasses, which tend to be merged together. with an added burr between the vertical support and the horizontal arm of the glasses.

[0006] Furthermore, prior art techniques are often limited to opaque and matte objects. A matte coating is generally used when the object is shiny or translucent, to avoid misinterpretations of the position of the object's outer surface during image processing or scanning. Without this coating, holes in the mesh obtained during a scan can form at surface reflections, particularly in the case of stereovision reconstruction with images taken from different angles (generally on the order of a few degrees). In addition, the internal parts that would be visible on the real object are then masked by the matte coating in the resulting images or scan, resulting in a loss of information for the rendering of the modeled object.

[0007] It should be emphasized that the three-dimensional representation of an object, obtained by prior art techniques, being a single block without segmentation, is not suitable for a virtual fitting of the modeled object, which requires articulation to better conform to a shape on which the object is positioned. For example, in the context of a virtual fitting of eyeglasses, or a superimposition of the three-dimensional representation of the object onto an image of the object in a video stream, the articulation and deformation of the temples relative to the front of the glasses may be necessary to obtain a realistic fit. In the case of a superimposition, alignment defects may also introduce biases, either when taking measurements or when masking the actual glasses visible in the image.These drawbacks are particularly significant in the context of augmented reality, where the processing of images from the video stream is performed constantly.

[0008] On the other hand, since the resulting three-dimensional representation is very large, capturing only the outer shell—that is, without knowledge of the internal parts—it becomes complex for a computer operator to easily process it in order to obtain a reliable segmented representation in terms of the architecture of external or even internal parts—for example, a metal core inside a translucent branch—or in terms of textures for each part. It frequently happens that the textures are imprecise, either in terms of texture type (solid or translucent material, plastic, glass, wood, etc.) or in terms of texture position, particularly with regard to details or inserts existing in the object being modeled. Similarly, a simplification of the model is often necessary to convert it into a model usable in real time.Existing meshing techniques in the state of the art are often resource-intensive and not optimized for the model to be reconstructed, since they have no prior knowledge of that model.

[0009] Since the resumption of the three-dimensional representation to obtain a satisfactory result is complex, it can hardly be carried out on a standard computer because it requires significant storage and computer memory capacities, or even a powerful graphics card to manipulate the computer file of the three-dimensional representation.

[0010] Furthermore, segmenting the resulting three-dimensional representation can prove tedious, particularly at the joints between several parts. It should be noted that joints are generally among the details smoothed in the three-dimensional representation obtained by conventional image processing and / or scans of the object.

[0011] Furthermore, with current modeling methods, the textures to be applied to the three-dimensional representation each have points grouped by proximity. It therefore happens that points located on two distinct parts end up on the same texture, making the texture difficult to correct by an external operator who has difficulty seeing at first glance which part(s) the texture corresponds to.

[0012] Finally, it regularly happens that the object to be modeled is deformed at the time of image acquisition or scanning, either by the mechanical constraints of the support, or by gravity as for example in the case of a handbag including a metal chain.

[0013] The three-dimensional representation then also includes the same deformations, which can lead to inaccuracies when the representation is used for taking a measurement, for example.

[0014] None of the current systems makes it possible to simultaneously meet all the required needs, namely to offer a modeling technique for an object from real data which makes it possible to obtain a three-dimensional representation of the object which is accurate and reliable in order to obtain a very good rendering of the object for an augmented reality application or for any other applications using a representation of a real object, in particular in terms of details and textures, and in particular for translucent parts of the object. Description of the invention

[0015] The present invention aims to remedy all or part of the disadvantages of the prior art mentioned above.

[0016] To this end, the invention relates to a method for modeling a real object, comprising the following steps: • Obtaining a two-dimensional and / or three-dimensional dataset, linked to the object, the dataset being associated with a first space; • Generation of a three-dimensional representation of the object in a second space from the dataset.

[0017] According to the invention, the three-dimensional representation comprises a plurality of parts whose surfaces are generated by a mathematical function describing a continuous parametric surface, the generation step of the three-dimensional representation of the object in the second, canonical space, comprising a sub-step of adjusting in shape and dimension each of the parts to at least one element of the real object, the adjustment being carried out taking into account at least one structural constraint of said object.

[0018] Thus, the rendering calculated on the three-dimensional representation of the real object is more accurate and less prone to rendering errors, because the surfaces of the parts composing the three-dimensional representation are continuous surfaces, calculated mathematically, for example, as a function of control points arranged in the second three-dimensional space. A mathematical function describing a mathematical surface via control points is, for example, a Bézier surface or non-uniform rational B-splines, also known as NURBS (an acronym for "Non-uniform rational B-spline"), composed of a plurality of Bézier or NURBS curves.Another example of a mathematical function describing a parametric surface is, for example, the concatenation of a plurality of elementary volumes, positioned relative to each other in the second three-dimensional space, each volume being parameterized by at least one value (for example, a radius for a sphere).

[0019] It should be emphasized that a mesh is a succession of points, each three-dimensional point being associated with a plurality of three-dimensional points, the set of points defining a discontinuous surface. The use of meshes in the modeling method according to the invention to describe the parametric surface can be considered but corresponds to a degraded mode of modeling the real object. It is possible to use both mathematical functions describing continuous parametric surfaces for the external or visible parts of the real object and mathematical descriptions defining mesh points delimiting a surface for internal or non-visible parts of the real object.

[0020] Furthermore, the accuracy of the rendering is also improved by the fact that the parts are modeled in a segmented manner, taking into account at least one constraint. Areas of detail can thus be modeled more finely than in the prior art, significantly increasing the quality of the rendering on translucent parts of the components, such as those with a metallic core. The surface normals are also smoother because they are derived from mathematical models and not from a discrete mesh (generated above a point cloud), which results in more pronounced reflections. continuous. Texture seams can also be placed in smaller areas if the semantics of the objects constituting the model are known, in order to limit visual artifacts.

[0021] A constraint can be internal to a part, such as a radius of curvature, the presence of a concavity, surface continuity, a relationship between two control points, or even continuity of normals at a point. Constraints can be purely experimental or derived from industry knowledge of the manufacturing process (for example, since most spectacle lenses are cut from a spherical cap, this shape constraint can be translated into a geometric constraint).

[0022] A constraint can also be of the inter-part type, such as an alignment relationship, an angular relationship, a connection between two parts, etc.

[0023] It should be noted that fitting a part can induce deformation of said part by modifying its dimensional parameters. Dimensional parameters can be global parameters – such as length, width, height, depth – or local parameters such as a radius of curvature or the position of a control point.

[0024] The three-dimensional representation being advantageously segmented, allows for finer modeling, which can be taken up more quickly by an operator who can effectively correct the defects of the initial dataset.

[0025] The augmented reality process using three-dimensional representation can thus have a better rendering, for example for a virtual try-on of glasses, be more precise, in particular when taking measurements in relation to the object, positioning the virtual object, masking the object or a virtual try-on.

[0026] Two-dimensional data from the resulting dataset can be, for example, an image of the object, a plan of the object, an artwork of the object, a map of the object, etc.

[0027] Preferably, two-dimensional data can be acquired by an imaging sensor, for example of the CMOS or CCD type. Two-dimensional data may advantageously be calibrated or not.

[0028] Three-dimensional data from the obtained dataset can be, for example, a three-dimensional model in a three-dimensional space, a 3D mesh, a point cloud in a three-dimensional space, etc. Three-dimensional data can be acquired by a three-dimensional imaging sensor, such as, for example, a scanner, an MRI or ultrasound imaging system, etc.

[0029] The initially obtained dataset can also be an elevation map, that is, two-dimensional data forming an image, each pixel of the image being associated with an elevation value, generally corresponding to a distance of the pixel from the image acquisition device. Such data, These data, subsequently referred to as 2.5D data, can include, for example, a depth map, a topographic map, or an architectural plan. A depth sensor can thus be used to acquire 2.5D data.

[0030] In particular embodiments of the invention, the fit of each part is constrained by the superposition of at least one curve on the surface of said part with a contour of a distinct element of the object, said contour being previously deduced from the data set.

[0031] In particular embodiments of the invention, a structural constraint is a predetermined geometric constraint on a part or between two parts.

[0032] Thus, a known architecture of the object can be followed, allowing for a significant improvement in the accuracy of the fit. For example, in the case of modeling a pair of glasses, it can be advantageous to know the composition of the tenon that acts as the hinge between the front of the glasses and the temple in order to accurately model this part, which tends to be merged during a scan of the object.

[0033] Geometric constraints can also be used to correct the three-dimensional representation in canonical space in order to compensate for any deformations of the object during data acquisition. Such constraints might include, for example, the requirement that the temples of a pair of glasses be parallel when open and form a 90° angle with the front of the glasses.

[0034] In particular embodiments of the invention, a structural constraint is a constraint of continuity of a curve of a part or a constraint of continuity of the tangent at at least one point of the curve of said part.

[0035] Thus, it is possible to automatically remove artifacts from a surface whose curvature is known. Such artifacts can arise, for example, from a fusion between a part of the object and the support, caused by a scan of the object.

[0036] In particular embodiments of the invention, the modeling process also includes a step of automatically determining a geometric transformation between the object in the acquisition space and the three-dimensional representation of said object in canonical space, the transformation comprising at least one of the following elements: • Straightening of a part of the three-dimensional representation according to a pre-established constraint on said part; • Adjusting an angle between two parts of the three-dimensional representation according to a pre-established constraint between said two parts; • Symmetrization of the three-dimensional representation.

[0037] Thus, the three-dimensional representation is of the canonical type, that is to say ideal with respect to a standard, because it is not distorted.

[0038] In particular embodiments of the invention, the modeling process also includes a step of identifying a contour of the object in the object dataset and a step of determining a surface curve of the three-dimensional representation, corresponding to said contour of the object.

[0039] In particular embodiments of the invention, the dataset includes data from an acquired image of the object, the modeling process also including a step of calculating a geometric deformation factor of at least one part of the object.

[0040] It should be emphasized that the geometric deformation can be a structural deformation, for example a temple of a pair of glasses adjusted by an optician to the morphology of an individual, or be an optical deformation related to an optical parameter of a system for acquiring said image, for example related to the focal length of a lens of the acquisition system.

[0041] In particular embodiments of the invention, the surface curve is an edge of the three-dimensional representation of the object.

[0042] In particular embodiments of the invention, the dataset includes at least two views of the object from distinct angles.

[0043] In particular embodiments of the invention, the dataset includes a view of the object and a depth map obtained concurrently with the view of the object.

[0044] In particular embodiments of the invention, the dataset comprises at least two depth maps obtained from distinct angles or a scan of the object.

[0045] In particular embodiments of the invention, the modeling process also includes steps of: • Recognition of at least one type of object shape and / or a part of the object in the dataset; • Automatic selection of the parametric three-dimensional model based on the previously recognized shape type(s).

[0046] In particular embodiments of the invention, the object is a pair of glasses, the recognition of at least one type of shape comprises all or part of the following sub-steps: • Recognition of a type of frame shape included in the pair of glasses; • Recognition of a type of lens shape included in the pair of glasses; • Recognition of a type of shape of a circle included in the pair of glasses; • Recognition of a type of shape of a tenon included in the pair of glasses; • Recognition of a type of shape of a bridge included in the pair of glasses; • Recognition of a type of shape of a hinge between the frame and a temple included in the pair of glasses; • Recognition of a type of temple shape included in the pair of glasses; • Recognition of a type of shape of a sleeve included on the branch; • Recognition of a plate shape type included in the pair of glasses.

[0047] In particular embodiments of the invention, the parts of the three-dimensional representation are derived from a generic three-dimensional model of the object, the relationships between said parts being inherited from said generic three-dimensional model.

[0048] For example, for modeling a sneaker, a generic shoe model is used. Such a generic model may include a sole, a heel, an upper, a vamp, a slider, a tongue, or even laces. The generic model is then adapted and adjusted to the initially obtained dataset.

[0049] The generic model thus comprises a collection of standard parts of a generic of the object to be modeled.

[0050] Advantageously, the generic model has been previously positioned and oriented in a similar manner to the object in a three-dimensional frame of reference of a virtual space representative of the object's environment.

[0051] For this purpose, a detection of characteristic points of the object may have been previously carried out in the dataset, the characteristic points allowing the generic three-dimensional model to be oriented and / or positioned in the three-dimensional frame.

[0052] Alternatively, the position and / or orientation of the object relative to a sensor that has acquired the dataset are known or estimated. In which case, the generic three-dimensional model can be positioned and / or oriented in the three-dimensional coordinate system relative to the sensor, whose position may, for example, correspond to the origin of the coordinate system.

[0053] In particular embodiments of the invention, the three-dimensional representation includes at least one articulation between two parts.

[0054] In particular embodiments of the invention, the generation of a three-dimensional representation is carried out in a time period of less than one second after obtaining a dataset.

[0055] The invention also relates to an electronic storage device for instructions of a modeling process according to any of the preceding embodiments.

[0056] Finally, the invention also relates to an augmented reality method implementing the method of modeling a real object according to any one of the preceding embodiments, and the resulting three-dimensional representation of said object. BRIEF DESCRIPTION OF THE FIGURES

[0057] Other advantages, purposes and particular features of the present invention will become apparent from the following non-limiting description of at least one particular embodiment of the devices and methods of the present invention, with reference to the accompanying drawings, in which: • [Fig. 1] is a synoptic diagram of an example of an implementation method of the modeling process according to the invention; • [Fig.2] is a photo of an example of a real object to be modeled, here a pair of glasses; • [Fig.3] is an image of a three-dimensional representation of the object of [Fig.3] obtained according to the process of [Fig.1]; • [Fig.4] is an image of a three-dimensional representation of the object in [Fig.2] according to a prior art technique; • [Fig.5] is a photo of another example of a real object to be modeled, namely here a shoe with laces; • [Fig.6] is an image according to a partial view of a three-dimensional representation of the object of [Fig.5] according to a technique of the prior art; • [Fig.7] is an image of a three-dimensional representation of an object similar to that of [Fig.5] according to the process of [Fig.1]; • [Fig.8] is a synoptic diagram of another example of an implementation method of the modeling process according to the invention. DETAILED DESCRIPTION OF THE INVENTION

[0058] The present description is given by way of non-limiting grammar, each feature of an embodiment being able to be advantageously combined with any other feature of any other embodiment.

[0059] It should be noted from the outset that the figures are not to scale. Example of a particular embodiment

[0060] Fig. 1 is a synoptic diagram of an example of an implementation method of a modeling method according to the invention.

[0061] The actual object 200 to be modeled in this non-limiting example of the invention, illustrated in [Fig.2] via an image obtained by an image acquisition device, is a pair of glasses comprising several elements including a translucent frame 210 revealing a metallic core 220 articulated by a hinge 225, allowing the front 230 and the temple 240 to be connected, through the tenon 250.

[0062] The method 100 comprises a first step 110 of obtaining a two-dimensional and / or three-dimensional dataset related to the object, the dataset being associated with a first acquisition space. For the sake of clarity in the illustration, the dataset is here two-dimensional, derived from photographs of the real object to be modeled.

[0063] This dataset is then processed in a second step 120 in order to generate a three-dimensional representation of the object in a second space called the modeling space.

[0064] The three-dimensional representation advantageously comprises a plurality of parts whose surfaces are generated by a mathematical function configured by control points. Such a mathematical function is, for example, a Bézier curve or NURBS. The notable advantage of generating surfaces using a mathematical function is obtaining a high-quality rendering regardless of the viewing angle or zoom level used, which is not possible with a conventional model obtained during an object scan because the discrete nature of conventional modeling introduces calculation errors in the rendering.

[0065] The curves of the three-dimensional representation are fitted during a substep 121 of step 120. This fitting is advantageously carried out taking into account the segmentation of the three-dimensional representation, which has been previously chosen to ensure the most accurate representation possible. The generation of the three-dimensional representation is thus performed by fitting a part of the three-dimensional representation to an element of the object to be modeled. It should be emphasized that the fitting of each part is constrained by the superposition of at least one surface curve of said part with a contour of a distinct element of the object, which has been previously deduced from the dataset.

[0066] The structural constraint can be a predetermined geometric constraint on a part or between two parts - for example a radius of curvature, a concavity, an alignment relationship between two points, etc. - or a constraint of continuity of a curve, or even of its normal.

[0067] More specifically, the adjustment is carried out here by means of projection onto a plane, then by a superposition of contours, one of which is derived from the data obtained Initially, one is the three-dimensional representation being calculated. An optimization calculation is then implemented, defining the parameters of the curves in the three-dimensional representation so that they fit as closely as possible to the contours detected in the initial dataset.

[0068] The modeling method 100 may also include a step 130 for automatically determining a geometric transformation between the object in the acquisition space and the three-dimensional representation in the second space so that the representation is canonical, i.e., undistorted, or even perfectly symmetrical in the case of a pair of glasses. Step 130 may be performed before, concurrently with, or even after step 120 to correct the geometric representation generated in step 120.

[0069] The geometric transformation used in step 130 generally includes at least one of the following elements: • Straightening a part of the three-dimensional representation according to a pre-established constraint on said part (different methods can be used for these transformations such as skeletonizing (also known by the English term "rigging") an animation skeleton or the use of a deformation field for example); • Adjusting an angle between two parts of the three-dimensional representation according to a pre-established constraint between said two parts; • Symmetrization of the three-dimensional representation.

[0070] The pre-established constraints can be, for example, angular constraints - a temple of glasses standing perpendicular to the face, a pantoscopic angle (orientation of the face of a pair of glasses with respect to the vertical) which is generally defined between 5 and 10 degrees, etc. - structural constraints - for example, straightening the outer lining of a bag which may have been deformed during the scan, placing a handle or chain of a bag in the direct extension of the bag, i.e. upwards when the bag is placed vertically - or even a symmetrization constraint in the case of an object whose architecture is symmetrical, as is the case of a pair of glasses.

[0071] For this purpose, the three-dimensional representation may advantageously include an articulation between two parts.

[0072] In order to allow the adjustment of the curves of the three-dimensional representation, before or concurrently with step 120, the modeling process 100 may also include a step 140 of identifying a contour of the object 200 in the dataset (by classical methods of filtering or machine learning for example) and a step 150 of determining a surface curve of the three-dimensional representation.

[0073] It should be noted that the modeling method 100 may also include a step 155 of calculating a geometric distortion factor related to an optical parameter of an image acquisition system for the real object. This distortion may, for example, be related to the focal length of a lens of the acquisition system. The induced distortion can be calculated beforehand using conventional techniques, in order to calibrate the camera and obtain optically undistorted images.

[0074] Advantageously, the modeling method 100 may also include a step 160 of recognizing at least one shape type of the object 200 or at least one shape type of a part of the object 200 in the dataset and a step 170 of automatically selecting the parametric three-dimensional model based on this recognition. Shape type recognition may, in particular, be performed by a conventional technique, based, for example, on landmark detection or on deep learning techniques.

[0075] Steps 160 and 170 allow the modeling process to be automated from a dataset.

[0076] Recognition step 160 may, in particular in the case of a pair of glasses, include recognition of the shape of a frame, of a circle, of a bridge, of a type of hinge, of a temple shape, or even of a nose pad shape.

[0077] It should be emphasized that the parts of the three-dimensional representation can be derived from a generic three-dimensional model of the object, for example, a model of a pair of glasses, shoes, or any other type of object. This allows for the preservation of intrinsic consistency of the object to be modeled, the type of which has been previously identified. The relationships between the parts can then be advantageously inherited from the generic three-dimensional model. The parametric three-dimensional model is a specification of the generic three-dimensional model, adapting the shape of each part to the object 200 to be modeled. A hierarchy of models can thus be established by progressively refining the shape of each part to arrive at a faithful three-dimensional representation of the real object 200. Statistical models such as a 3DMM (acronym for "3D Morphable Model") can also be used for these purposes.

[0078] The [Fig.3] is an image of the three-dimensional representation 300 of the object 200 obtained by the modeling process 100.

[0079] Compared with [Fig. 4], which is an image of a three-dimensional representation 400 of the same object 200 but obtained by a conventional prior art technique, namely a three-dimensional scan of the object 200, the three-dimensional representation 300 appears significantly more precise than the Three-dimensional representation 400. In particular, unlike three-dimensional representation 400, three-dimensional representation 300 has a very smooth surface and very precise details. The part 310 representing the metal core 220 is thus well positioned in three-dimensional representation 300, inside the envelope 320 of three-dimensional representation 300, whereas the metal core 220 is formed in three-dimensional representation 400 by a discontinuous texture 420 applied to the outer surface.

[0080] In addition, protrusions 410 related to the support or bumps 430 are visible on the three-dimensional representation 400 obtained by the prior art technique, which is not the case of the three-dimensional representation 300 which is advantageously smoothed by the structural constraints implemented during the modeling process 100.

[0081] Once the modeling is complete, the three-dimensional representation can advantageously be implemented within an augmented reality process, for example by overlaying a video stream onto a real image. A virtual try-on of an object can thus be implemented more realistically because the resulting rendering has fewer negative visual effects. Furthermore, the modeling process allows for the segmentation and semantics of each part of the modeled object, which is necessary for the correct positioning of these objects in an augmented reality context. For example, the nose pads will be easily identified for placement on the nose. Other advantages and optional features

[0082] An object to be modeled is included from the following non-exhaustive list: • a pair of glasses; • a pair of rimmed glasses; • a pair of semi-rimless glasses; • a pair of rimless glasses; • a pair of multi-rimmed glasses; • a pair of one-piece glasses. • a pair of full-rimmed glasses with a single bridge; • a pair of full-rimmed glasses with a double bridge; • a pair of semi-rimless glasses with a single bridge; • a pair of semi-rimless glasses with a double bridge; • a pair of goggle-type glasses • a bag; • a shoe, or even a pair of shoes; • a piece of jewelry; • a watch; • a helmet; • an item of clothing; • a piece of furniture.

[0083] Generally speaking, an object to be modeled is a manufactured object. However, the The modeling process can also be applied to a part of a human or animal body, such as a head.

[0084] Fig. 5 is a photo of a shoe 450 with laces 455 to be modeled.

[0085] Figure 6 is an illustration of part of a three-dimensional representation 460 of shoe 450 in [Fig. 5] according to a prior art method, namely a scan. The laces 455 of the model 460 are fused together. It is also difficult to delineate the laces 455, the tongue 465 and the two quarters 466 of shoe 450 and consequently to easily replace them.

[0086] By using the modeling process 100 illustrated in [Fig.1] on a similar shoe, the three-dimensional representation 470 illustrated in [Fig.7] is much more consistent, modifiable and usable by clearly separating each lace 475, the tongue 476 and the quarters 477 of the shoe 478. Another detailed example of implementation

[0087] Figure 8 illustrates an example of implementation of a modeling method 500 according to the invention, in the form of a synoptic diagram.

[0088] The modeling process 500 initially includes a step 510 of obtaining 2D or 3D input data, from photos, plans, 3D scans, etc. of the real object 200.

[0089] A general qualification of the real object 200 to be modeled is carried out initially during a step 515. This qualification includes in particular a recognition of the type of the object 200 by classification tools well known to the person skilled in the art.

[0090] A geometric transformation is then evaluated in a step 520 to correct the deformations undergone by the real object 200 in order to rectify it in a perfect canonical space. It should be noted that the geometric transformation can be reversible.

[0091] The data are then positioned in whole or in part in the second three-dimensional space, called canonical, during a step 530, taking into account any known structural constraints of the object, for example positioning of a branch at 90° with respect to a face of a pair of glasses.

[0092] A parametric model, forming all or part of the three-dimensional representation of object 200, is then selected from a database during a step 540, to model at least a part of object 200. An order particulars of the parts to be modeled can be followed according to the architecture of the previously qualified object 200.

[0093] This selection is carried out by identifying characteristic points or characteristic landmarks in different views and then by matching them with a parametric model.

[0094] The parametric model is positioned and deformed according to the transformation determined in step 520 in the second space so that it overlaps with the initial data during a step 550, in order to maintain the consistency of the object 200.

[0095] Advantageously, a hinge of the parametric model can be advantageously positioned relative to a hinge recognized in the initial data. Similarly, a bridge or plates can also be selected and positioned.

[0096] In the case where the initial data are images of object 200, the image of the object may have undergone distortions due to the optics used. Optical distortion can also be calculated and taken into account in order to correct the images of object 200 and limit errors in the interpretation of the initial data.

[0097] It should be emphasized that the selection and default positioning of the parametric model are here carried out automatically by the use of a general parametric representative model which is fitted on different views of the object 200. For example, for a pair of glasses, the general parametric model is a statistical model which is deformable on the one hand and contains a hierarchy and a set of relations - positioning constraints, orientation constraints, animation constraints - between the constituent elements each represented by a proper parametric model.There may be several standard models corresponding to a subclass of objects: for example, the general model of a boot is different from that of a pair of sneakers even though they belong to the same general class of shoes, the general class defining a more generic model made up of elements having a semantics related to the construction of the manufactured object, namely a sole, a seam, a tongue, a closure, heels, etc.

[0098] An adjustment of the shape of the parametric model is then carried out during a step 560 by optimizing a superposition of curves projected onto a plane, curves from the parametric model and the initial data.

[0099] The adjustment of the parametric model is carried out via intrinsic parameters (control points, size, etc.) and / or extrinsic parameters (position, orientation, torsion directions, etc.). Positioning constraints can advantageously be taken into account.

[0100] A parametric model is a collection of curves and surfaces that have editing or adjustment constraints according to business parameters with a Mathematical parameter matching. This limitation of the space of possible deformations of an element of the parametric model ensures that this element is always represented with constraints that may be necessary given the object's architecture. The business logic is thus directly integrated into the modeling. A parametric model also contains positioning and interaction elements with the representation of the standard parametric model and other constituent elements, which guarantees its placement and orientation in the construction of the manufactured object. Thus, a hinge on the temples of eyeglasses is always in the correct position and always connects the temple to the frame of the glasses.

[0101] Steps 540 to 560 can be repeated in order to model each part, element or piece of object 200.

[0102] It should be emphasized that the adjustment performed during step 560 can be repeated for a plurality of images taken from potentially different viewpoints, in order to finalize the shape of the parametric model. Depending on the image (or more generally on the type of data), a geometric transformation can be applied to rectify the image of object 200 so that it can be used as a model for the three-dimensional representation in the second canonical space.

[0103] Finally, an operator can manually adjust the three-dimensional representation of the object 200, formed by each parametric model, in order to perfect the three-dimensional representation during a step 570. It should be emphasized that this step 570 is optional.

Claims

Demands

1. A method for modeling a real manufactured object comprising a plurality of elements, comprising the steps of: • Obtaining a two-dimensional and / or three-dimensional dataset related to the object, the dataset being associated with a first space; • Generating a three-dimensional representation of the object in a second space from the dataset; characterized in that the three-dimensional representation comprises a plurality of parts whose surfaces are generated by a mathematical function describing a parametric surface, the parts of the three-dimensional representation being derived from a generic three-dimensional model of the object, the relationships between said parts being inherited from said generic three-dimensional model, one relationship being included in the list {positioning constraint, orientation constraint, animation constraint},The step of generating the three-dimensional representation of the object in the second, canonical space, comprising a sub-step of adjusting the shape and dimensions of each part to at least one element of the real object, the adjustment being carried out taking into account at least one structural constraint of said object, the process also comprising a step of automatically determining a geometric transformation between the object in the acquisition space and the three-dimensional representation of said object in canonical space, the transformation comprising at least one of the following elements: • Straightening a part of the three-dimensional representation according to a pre-established constraint on said part; • Adjusting an angle between two parts of the three-dimensional representation according to a pre-established constraint between said two parts; • Symmetrizing the three-dimensional representation.

2. A modeling method according to claim 1, wherein the fit of each part is constrained by the superposition of at least one curve on the surface of said part with a contour of a a distinct element of the object, said outline being previously deduced from the dataset.

3. A modeling method according to any one of claims 1 to 2, wherein a structural constraint is a predetermined geometric constraint on a part or between two parts.

4. A modeling method according to any one of claims 1 to 3, wherein a structural constraint is a continuity constraint of a curve of a part or a continuity constraint of the tangent at at least one point of the curve of said part.

5. A modeling method according to any one of claims 1 to 4, also comprising a step of identifying an object contour in the object dataset and a step of determining a surface curve of the three-dimensional representation, corresponding to said object contour.

6. A modeling method according to any one of claims 1 to 5, wherein the dataset includes data from an acquired image of the object, the modeling method also comprising a step of calculating a geometric deformation factor of at least one part of the object.

7. A modeling method according to any one of claims 1 to 6, wherein the surface curve is an edge of the three-dimensional representation of the object.

8. A modeling method according to any one of claims 1 to 7, wherein the dataset comprises two views of the object from distinct angles.

9. A modeling method according to any one of claims 1 to 8, wherein the dataset includes a view of the object and a depth map obtained concurrently with the view of the object.

10. A modeling method according to any one of claims 1 to 9, wherein the dataset comprises at least two depth maps obtained from distinct angles or a scan of the object.

11. A modeling method according to any one of claims 1 to 10, comprising the steps of: • Recognition of at least one type of shape of the object and / or a part of the object in the dataset; • Automatic selection of the parametric three-dimensional model based on the previously recognized type(s) of shape.

12. A modeling method according to claim 11, wherein, where the object is a pair of glasses, the recognition of at least one shape type comprises all or part of the following substeps: • Recognition of a shape type of a frame included in the pair of glasses; • Recognition of a shape type of a circle included in the pair of glasses; • Recognition of a shape type of a tenon included in the pair of glasses; • Recognition of a shape type of a bridge included in the pair of glasses; • Recognition of a shape type of a hinge between the frame and a temple included in the pair of glasses; • Recognition of a shape type of a temple included in the pair of glasses; • Recognition of a shape type of a sleeve included on the temple; • Recognition of a shape type of nose pad included in the pair of glasses.• Recognition of a type of logo shape or inserts included in the pair of glasses.

13. A modeling method according to any one of claims 1 to 12, wherein the three-dimensional representation includes at least one articulation between two parts.

14. A modeling method according to any one of claims 1 to 13, wherein the generation of a three-dimensional representation is carried out in a time period of less than one second after obtaining a dataset.

15. Electronic storage device for instructions of a modeling process according to any one of claims 1 to 14. 20

16. Augmented reality method implementing the method of modeling a real object according to any one of claims 1 to 14, and the three-dimensional representation of said object obtained.

Images