METHOD AND SYSTEM FOR VIRTUAL PREDICTION OF THE REAL SEAT OF A REAL EYEGLASS FRAME ON THE HEAD OF A PERSON WITH INDIVIDUAL HEAD GEOMETRY

Description

[0001] The present invention lies in the field of spectacle technology and relates to a method, a system, and a computer program for the virtual prediction of the actual fit of a real spectacle frame of a specific frame geometry on the head of a person with an individual head geometry. The invention can be used in particular within the framework of a virtual try-on, hereinafter also abbreviated as "VTO" (from English "Virtual Try-on").

[0002] One goal of such a VTO (Virtual Tour Operator) can be, in particular, to provide the person with a plausible preview of the eyeglass frame and, above all, its fit on their head, in a scenario where an eyeglass frame does not yet exist or at least is not yet physically present in the person's possession (hereinafter also referred to as "person"). This preview can be achieved by means of a virtual representation of the eyeglass frame (3D model) on the one hand and the person's head (or a part of the head) on the other. Such a preview can be designed, in particular, as a video, image, or even a 3D model.

[0003] In the prior art, three main methods for virtual product demonstration (VTO) are known, which differ primarily in their visual representation: Firstly, there are AR-based systems (AR = Augmented Reality) in which a 3D product is superimposed in real time onto a live video stream from a smartphone or webcam. Such a preview generally appears realistic, as the overall image consists largely of the video showing the real person, while the synthetic, rendered part of the image showing the product often occupies only a small portion of the frame.

[0004] Secondly, there are also VTO variants where the entire preview image is fully synthesized / rendered using established computer graphics techniques. For this to work, in addition to the 3D object being previewed, the person's body region (in the case of glasses, the person's head) must also be at least partially available as a 3D model. As in a video game, the preview then consists entirely of virtual objects. To create the 3D model of the body region, a 3D scan of the region is typically performed before such a preview. Once the body region is digitized as a 3D model, it can be arranged three-dimensionally in virtual space together with the product, which already exists as a 3D model. The rendering pipeline of a real-time computer application then creates a smooth, real-time image, just like in a video game.

[0005] Thirdly, hybrid VTO systems now exist, in which a 3D model of the person is created or generated either before or during an AR application. This 3D model of the person, or a part of their body, is then not visible to the VTO user (i.e., the person). Instead, they only see a live video stream of themselves and the 3D model of the product displayed relative to it. In the VTO process, the person's 3D model is used, for example, to correctly determine or render hidden areas of the product model that is currently being tried on in the VTO.

[0006] US Patent 2021 / 0065285 A1 describes systems, methods, and machine-readable media for the virtual fitting of items such as eyeglasses and / or eyeglass frames. A user interface for virtual fitting can be implemented on a server or a user device and uses three-dimensional information for the user and three-dimensional information for each frame, with the frame information stored in a frame database to identify and / or recommend frames that are likely to fit the user. The fit information can be provided for a group of eyeglass frames or for each individual frame selected by the user. The fit information can be provided with a static image of the eyeglass frames and / or as part of a virtual try-on, in which the eyeglass frames are virtually placed on a real-time image of the user.

[0007] EP 3 410 178 A1 describes methods, computer programs, and devices for virtual spectacle try-on. For this purpose, 3D models of a head and a spectacle frame, as well as head metadata based on the head model and frame metadata based on the frame model, are provided. The head metadata includes contact information, in particular a contact point, which can be used for the initial positioning of the spectacle frame on the head, and / or a contact area, which describes an area of ​​the frame's temples for resting on the ears of the head.

[0008] It is an object of the present invention to further improve the method of generating a prediction of the actual fit of a real eyeglass frame on the head of a person with individual head geometry, particularly within the framework of a VTO (Virtual Trial Opt-Out). In particular, it is desirable to achieve an improvement with regard to increased efficiency and / or real-time capability of a computer-aided method used to generate the prediction, given available computing resources.

[0009] The solution to this problem is achieved according to the teaching of the independent claims. Various embodiments and further developments of the invention are the subject of the dependent claims.

[0010] A first aspect of the invention relates to a method, particularly computer-implemented, for the virtual prediction, especially in the context of a virtual fitting, of the real fit of a real eyeglass frame of a specific frame geometry on the head of a person with individual head geometry, wherein the method comprises: (A) Acquisition of head model data representing a three-dimensional virtual head model of at least part of the person's head; (B) Acquisition of spectacle model data representing a three-dimensional virtual spectacle model of at least part of the frame geometry of the spectacle frame;(C) A first placement step in which a set of points from a plurality of spectacle reference points predefined by or based on the spectacle model data on the spectacle model is aligned relative to a set of points from one or more initial head reference points predefined by or based on the head model data on the head model, in order to define an initial virtual placement of the spectacle model on the head model. The aforementioned sets of points may be defined, in particular, by metadata associated with the respective spectacle reference points or head reference points, which describe for each of these points to which sub-area of ​​the spectacle model or the head model the respective point belongs. For example, one or more initial head reference points located at the bridge of the nose of the head model ("bridge of the nose points") and the initial spectacle reference points associated with them may each be defined by the metadata;(D) an adjustment process with one or more further placement steps iteratively performed following the first placement step, in each of which, by means of raycasting, one or more additional head reference points are determined on the head model using a respective virtual ray (half-line) emanating from one or more of the spectacle reference points and are used to determine at least one parameter value of a parameterized translational and / or rotational transformation of the spectacle reference points for adjusting the virtual placement of the spectacle reference points relative to the head model and the correspondingly defined transformation with respect to the spectacle reference points is executed;and (E) Generating and outputting predictive information which, based on the virtual placement of the eyeglass reference points relative to the head reference points resulting from the transformation of the last executed placement step during the iteration, represents a prediction of the actual position of the real eyeglass frame on the person's head.

[0011] The term "capture" of head model data or spectacle model data, as used herein, refers in particular to (i) accessing such data, for example by receiving the data from a data source, e.g. via a communication link or other data interface (e.g. user interface) or by reading from a data storage device, or (ii) generating such data within the process itself, in particular by means of sensory capture (especially by 3D scanning) or calculations based on a computer model depending on input data that represents information characteristic of the shape of the person's head or the spectacle frame.

[0012] The terms "aligned" or "aligning," etc., of two sets of points, as used herein, mean in particular that at least one of the two sets of points, here especially the set of spectacle reference points, is moved in virtual space, in particular by translation and / or rotation, in order to bring it into a predetermined spatial relationship to the other set of points with respect to at least one available degree of freedom (normally, unless boundary conditions prevent this, there are three translational and three rotational degrees of freedom in three-dimensional space). In doing so, all points of a set of points to be moved are subjected to the same transformation, so that within each set of points the relative arrangement of the points is transformation-invariant, i.e., it is preserved during the transformation.In particular, the predetermined spatial relationship can be defined such that one or more points of one set of points are brought into alignment with corresponding points of the other set of points by means of the transformation(s).

[0013] The term "virtual," as used herein, is to be understood in particular as meaning that the "virtual" object or action thereby characterized relates to a reality simulated by a computer. A virtual object, such as a head model or eyeglass model within the meaning of the invention, therefore does not exist as such in reality, but only as an object in virtual space. Thus, when it is stated that the eyeglass model is virtually placed on the head model, this means that in virtual space a virtual eyeglass frame, represented by the eyeglass model, is placed, i.e., arranged, on the virtual head, represented by the head model. The same applies when it is stated, in abbreviated form, that the eyeglass model is placed on the head model.

[0014] The term "glasses model," as used here, refers to a virtual representation of at least part of a pair of glasses. Similarly, the term "head model," as used here, refers to a virtual representation of at least part of a person's head. Each of these models can be defined, in particular, as a 3D mesh, specifically a polygon model (polygon net), with nodes and edges.

[0015] The term "head reference points," as used herein, refers in particular to specific selected points on the head model that serve as head-side reference points for aligning the spectacle model or its spectacle reference points on the head model or its head reference points. Similarly, the term "spectacle reference points," as used herein, refers in particular to specific selected points on the spectacle model that serve as frame-side reference points for aligning the spectacle model with the spectacle reference points on the head model or its head reference points.

[0016] The term "raycasting" (usually spelled "ray casting" in English), as used here, refers specifically to an algorithm for calculating occlusion—that is, determining the visibility of three-dimensional objects from a specific point in space—based on the emission of virtual rays in virtual space. Starting from a point of origin, in particular a reference point for the glasses, a virtual ray is "emitted" along a straight line in a predetermined direction, and, if necessary, an intersection point of the ray with a scene, specifically a surface of the head model, is determined. Such a specific point can, in particular, be defined as an (additional) head reference point.In particular, two antiparallel rays can be emitted from the starting point, allowing it to be determined, especially based on the position of a defined intersection point with a surface of the head model, whether the starting point is located inside or outside the head model. Occasionally, in the technical field of computer graphics, the term "ray tracing" is used synonymously with "raycasting," although the term "raycasting" can also have other meanings in computer graphics.

[0017] The term "prediction," as used herein, refers in particular to an estimate or calculation of information describing a quality or quantity from which conclusions can be drawn about a good or poor fit, especially wearing comfort, stability, or a vision-enhancing or vision-impairing relative positioning of the spectacle model on the head model, or correspondingly, of a real spectacle frame modeled in the spectacle model on the real head of a real person modeled in the head model. In particular, the prediction may include a visual representation of the resulting fit of the spectacle model on the head model or one or more pieces of information derived therefrom, especially in the sense of a quality measure for the fit.

[0018] Any terms used herein, such as "comprises," "includes," "includes," "has," "with," or any other variant thereof, are intended to cover non-exclusive inclusion. For example, a procedure or system that comprises or has a list of elements is not necessarily limited to those elements but may include other elements not expressly listed or inherent in such a procedure or system.

[0019] Furthermore, unless explicitly stated otherwise, "or" refers to an inclusive or and not an exclusive "or". For example, a condition A or B is satisfied by any of the following: A is true (or present) and B is false (or not present), A is false (or not present) and B is true (or present), and both A and B are true (or present).

[0020] The terms "ein" or "eine," as used here, are defined as "one or more." The terms "ein anderer" and "ein Weitere," as well as any other variant thereof, are to be understood as "at least one more."

[0021] The term "plural", as it may be used here, is to be understood in the sense of "two or more".

[0022] The terms "configured" or "set up" to perform a specific function (and any variations thereof), as used herein, are to be understood, within the meaning of the invention, as meaning that the corresponding device or system already exists in a configuration or setting in which it can perform the function, or at least is adjustable—i.e., configurable—so that it can perform the function after appropriate adjustment. Configuration can be achieved, for example, by adjusting parameters of a process sequence or by using switches or similar devices to activate or deactivate functionalities or settings. In particular, the device / system can have several predetermined configurations or operating modes, so that configuration can be carried out by selecting one of these configurations or operating modes.

[0023] Using the method described in the first aspect, the actual fit of the real eyeglass frame, modeled in the eyeglass model, to the real head of the person, modeled in the head model, can be determined with high accuracy. This includes, in particular, the physical interaction and collision detection between the two models, allowing conclusions to be drawn about a good or poor fit. Specifically, this method allows for statements about the correct fit of eyeglasses relative to the wearer's face, i.e., the person, in virtual space, even before the real eyeglass frame exists.

[0024] Since the method relies primarily on a relative positioning of the spectacle reference points to the head reference points, especially to the additional head reference points, without requiring consideration of the entirety of both models, particularly all nodes and / or edges of polygon models, to determine the fit of the spectacle model on the head model, the method can be executed faster with relatively little computational effort and thus even with limited computing resources that would be overwhelmed by a conventional VTO method of the same quality (e.g., in many of today's battery-powered mobile devices, especially smartphones or tablet computers), up to real-time execution.In this way, a virtual reality trial (VTO) based on the first aspect of the process can quickly and efficiently provide the future eyeglass wearer with a reliable prediction of the expected comfort, aesthetics, and fit of the real eyeglass frame, virtually represented by the eyeglass model, on their head. This can therefore help to avoid incorrect purchases and support the selection of the optimal eyeglass frame for the individual.

[0025] Preferred embodiments of the method are described below, which, unless expressly excluded or technically impossible, can be combined with each other and with the other described aspects of the invention as desired.

[0026] To illustrate some of the embodiments, a local Cartesian coordinate system anchored in the spectacle model is used below. As is common with spectacles, the spectacle model has a front part with the respective frames for two lenses, a nose bridge connecting the two frames, and two earpieces attached to the front part. The origin of the local coordinate system lies at a central point on the nose bridge. If a line is drawn connecting the respective geometric centers of the frame for the right eye and the frame for the left eye of a (virtual) spectacle wearer, or the head model in a normal spectacle fit on the head or head model, then this connecting line defines the X-direction of the local coordinate system. The gaze direction of the (virtual) spectacle wearer according to the head model is defined here, by way of example, as the Z-direction of the local coordinate system.The Z-direction can also be defined, specifically in relation to the spectacle frame itself, as running parallel to the fully extended temples of the spectacle frame before it is fitted. However, these two exemplary definitions are generally equivalent. The Y-direction runs perpendicular to the X-direction and the Z-direction in such a way that a right-handed coordinate system results, with the X-axis as the first axis, the Y-axis as the second axis, and the Z-axis as the third axis.

[0027] The initial placement according to the first placement step can therefore be expediently carried out in such a way that, as a result, the Z-axis of the local coordinate system (defined in relation to the eyeglass model) coincides with the straight-ahead gaze direction of the (virtual) eyeglass wearer according to the head model. An example of such a local coordinate system is shown in Fig. 1 illustrated. As part of the subsequent adjustment process, the position of the spectacle model relative to the head model can, however, be changed in some cases so that the Z-direction of the local coordinate system anchored in the spectacle model and the straight-ahead viewing direction of the (virtual) spectacle wearer according to the head model no longer coincide.

[0028] In some embodiments, determining the transformation for the current placement step—that is, the step currently being traversed within the method—involves adjusting at least one parameter of the transformation depending on the relative position, in particular a geometric distance between at least one of the spectacle reference points and at least one of the additional head reference points determined in the current placement step, in order to achieve an adjustment target defined with respect to this position. For example, it might be determined during the transformation process that a particular spectacle reference point is spaced away from the surface of the head model, even though, according to an adjustment target, it is actually intended to rest on this surface.In this case, at least one relevant parameter can be determined as a function of this distance such that the transformation maps or shifts the spectacle reference point onto the surface. On the other hand, it might be discovered during the process of determining the transformation that a particular spectacle reference point lies within the head model, which in the real world would correspond to a state in which a corresponding section of the spectacles has "penetrated" the person's head. To counteract such an obviously undesirable or even pointless scenario, and to modify the position of the spectacle model on the head model to avoid this scenario, at least one relevant parameter of the transformation can now be determined such that the spectacle reference point is transformed onto the surface of the head model if it is intended to be located there (adaptation goal 1), or otherwise, at a distance from it, into the free space in front of it (adaptation goal 2).As already mentioned, in all transformations used in the procedure, never a single spectacle reference point, but always the entire set of spectacle reference points is subjected to the respective transformation, so that the relative arrangement of the spectacle reference points among themselves remains invariant during the transformations. This means, in particular, that the positions of the spectacle reference points in the local coordinate system are invariant.

[0029] In some embodiments, the first placement step, or at least one of the placement steps of the fitting process, further comprises: (i) determining at least one additional head reference point on the head model for each side of the head by means of raycasting based on a virtual ray (half-line) emanating from a respective spectacle reference point on an earpiece of the spectacle model corresponding to the same side of the head; and (ii) adjusting at least one parameter of the transformation depending on the respective relative position, in particular a respective distance, of each of these spectacle reference points to the additional head reference point determined from it, in order to achieve an alignment of the spectacle model that is equidistant with respect to the left and right sides of the head model as the fitting objective.The parameter of the transformation can be chosen, in particular, to define a rotation around the Y-axis, for example as a rotation angle.

[0030] In some embodiments, at least one of the placement steps of the adaptation process for determining the transformation for that placement step comprises one or more of the following subprocesses, in particular according to the following order: (a) Adjusting at least one parameter of the transformation as a function of the relative position, in particular a distance, of a head reference point, in particular the ear root point, on one ear of the head model to a corresponding spectacle reference point on an earpiece of the spectacle model associated with the ear, in order to achieve, as the adjustment goal, a placement of the spectacle model on the head model in which the earpiece rests on the ear root of the ear. The adjustment can, in particular, also be carried out accordingly with respect to both ears. The parameter(s) of the transformation can, in particular, be chosen such that a rotation about the X-axis and / or the Z-axis is defined, for example, as the respective rotation angle for each axis.for example, in the case of a combined rotation first about the X-axis and then about the Z-axis; (b) determining at least one additional head reference point on the nose of the head model by means of raycasting in a direction at least partially parallel to the anatomical horizontal axis of the head model, using a virtual ray (half-line) emanating from a respective spectacle reference point on a nasal area of ​​the spectacle model configured to at least partially encompass the nostrils of a nose of the head model; and adjusting at least one parameter of the transformation depending on the respective relative position, in particular a respective distance, of each of these spectacle reference points to the additional head reference point determined from it.The goal of the adaptation is to center the nose area of ​​the spectacle model on the head model with respect to the direction parallel to the anatomical horizontal axis. The parameter(s) of the transformation can be chosen, in particular, to define a rotation about the Y-axis, for example, as a specific angle of rotation around the Y-axis. The Y-axis can advantageously be positioned such that it intersects the X-axis at midway along the straight line connecting the two ear reference points, i.e., at the center of the head.(c) Determining at least one additional head reference point on the nose pad in the X direction, but hardly affecting the earpiece placement on the ear points; (c) Determining at least one additional head reference point on the nose of the head model by means of raycasting in a direction at least partially parallel to the anatomical longitudinal axis of the head model, using a virtual ray (half-line) emanating from a respective spectacle reference point on a nasal area of ​​the spectacle model configured to at least partially encompass the nostrils of a nose of the head model; and adjusting at least one parameter of the transformation depending on the respective relative position, in particular a distance, of each of these spectacle reference points to the additional head reference point determined from it.to achieve, as an adjustment goal, at least partial contact of the nose area of ​​the spectacle model with the nostrils of the nose of the head model. The parameter(s) of the transformation can be chosen, in particular, such that it defines a rotation about an axis parallel to the X-axis through the earpieces, in particular about an axis of rotation defined as a line connecting the ear root points of both sides of the head model, for example, as the respective angle of rotation about such an axis; (d) Check whether, according to the current placement of the spectacle model on the head model, there is a collision of the spectacle model with a cheek area of ​​the head model; and if this is the case, adjust at least one parameter of the transformation to achieve, as an adjustment goal, the elimination of the collision with a cheek area. The parameter(s) of the transformation can be chosen, in particular, such thatthat this defines a rotation about an axis parallel to the X-axis through the earpieces, in particular about an axis of rotation defined as a straight line connecting the ear root points on both sides of the head model, for example as the respective angle of rotation about such an axis opposite to the direction of rotation from step (c); (e) Check whether, according to the current placement of the spectacle model on the head model, there is a collision of the spectacle model with a forehead or bridge of the nose area of ​​the head model; and if this is the case, adjust at least one parameter of the transformation in order to achieve, as the adjustment goal, the elimination of the collision with the forehead or bridge of the nose area. The parameter(s) of the transformation can, in particular, be chosen such that a translation along a specific,The vector referred to here as the temple vector is defined. The temple vector can, in particular, result from an orthogonal projection of a direction vector connecting the starting point and the endpoint of an earpiece onto the local Y / Z plane of the spectacle model.

[0031] The term "ear root point", as used herein, refers to a point, in particular the uppermost point in the anatomically longitudinal direction, at the root of an ear on the head model or correspondingly on the real head of the person.

[0032] In some embodiments, the method further includes a displacement process carried out between the first placement step and the adjustment process, in which the spectacle reference points are subjected to a translational transformation in a direction along or parallel to a nasal bridge line of the head model, which extends between a head reference point located at the nasal root of the head model and another head reference point located on the nasal bridge, in particular at the nasal tip, of the head model, in particular in a straight line.As part of the adjustment process, user input can be requested, allowing the user to specify a different placement point for the glasses model on the bridge of the nose in the head model. This enables the system to shift this placement point, particularly towards the tip of the nose, thus indicating to the process how far the user wants their glasses to be worn from their forehead. Alternatively, information about the desired placement point can be provided to the process in advance as user-specific configuration data, without requiring user input.

[0033] In some embodiments, the method further includes a fit determination process following the adaptation process, which comprises at least one of the following sub-processes (i) to (vi) (including any other sub-processes that may be required): (i) Determine, based on the placement of the spectacle model on the head model resulting from the fitting process and for at least one earpiece of the spectacle model, the respective length of an earpiece segment extending between a front part of the spectacle model and the point of contact of the earpiece at an ear root point on the head model. This can be used, in particular, to infer the fit of the earpiece when its total length is known, especially whether it is too long or too short and therefore offers a suboptimal fit on the ear. (ii) Estimate the contact pressure of an earpiece of the real spectacle frame corresponding to the earpiece of the head model on the person's real head by calculating a virtual contact pressure of the earpiece of the spectacle model on the head model as a function of the determined length of the earpiece segment and known elastic properties of the real earpiece.This can be used in particular to infer the fit of the temple in the lateral direction and thus also an important aspect of the wearing comfort of a real spectacle model corresponding to the spectacle model relative to the person's real head corresponding to the head model. (iii) Determine, based on the placement of the spectacle model on the head model resulting from the fitting process and for at least one sub-area of ​​the spectacle model that has several spectacle reference points, what proportion of these spectacle reference points rest on the head model. This proportion can be determined in particular absolutely as a number or relatively, e.g. as a percentage. This can be used in particular to infer the fit of the spectacle model and thus of the corresponding real spectacle frame as a whole or individually of at least one sub-area on the head model or the corresponding real head of the person.For the spectacle frame as a whole, or at least the relevant sub-area, corresponding reference fit data can be defined or created. These data can be compared with the proportion(s) determined by sub-process (iii) in order to draw conclusions about the quality of the fit of the spectacle model or the corresponding real spectacle frame, particularly according to one of the sub-processes (iv) or (v) mentioned below. It is also conceivable to make a decision, depending on the proportion, as to whether the fitting process was successful or not, and in the latter case, to terminate the process (and optionally issue an error message) or to repeat it.(iv) Determine, based on the placement of the spectacle model on the head model resulting from the fitting process, the relative spatial position of at least one selected spectacle reference point with respect to at least one point corresponding to a head reference point, or determined depending on the respective positions of several head reference points. For example, a distance can be calculated from spectacle reference points on the upper and lower front of the spectacle frame that at least approximately describes the height of the spectacle frame. Based on the respective positions of head reference points defined at the eyes, in particular the pupils, a viewing point of the spectacle wearer can then be determined, at least roughly, for each eye or lens opening in the spectacle frame. The viewing point should ideally be located, for example, slightly above the geometric center of the corresponding lens opening in the spectacle frame.a spectacle lens lies within it. (v) Estimating the quality of the fit of the real spectacle frame on the person's real head as a function of the contact pressure estimated (according to subprocess (ii)), the proportion of the spectacle reference points of the sub-area in contact with the head model determined (according to subprocess (iii)), and / or one or more relative positions determined (according to subprocess (iv)); (vi) Determining a fit rating by comparing the quality of the fit estimated (according to subprocess (v)) with corresponding reference data; (vii) Outputting the determined quality and / or rating of the fit as part of the prediction information. The output can be provided, in particular, via a user interface to inform a user of the result of the fit rating. Additionally or alternatively, the output can also include providing output data representing the fit rating or a quantity derived therefrom.In particular, depending on this output data, a subsequent process can be initiated, configured, or controlled, such as a process for creating a (modified or refined in terms of its resolution) 3D model of the eyeglass frame for its subsequent custom manufacturing, especially using a 3D printing process.

[0034] In the case of an application of the measurement parameter determination process described below, the information determined in one or more of the above sub-processes (i) to (vii) can also be used as input for the measurement parameter determination process in order to effect a correspondingly adapted form of the spectacle model and the real spectacle frame to be subsequently manufactured as custom-made spectacles depending on it.

[0035] Accordingly, in some embodiments, the method further includes a dimension parameter determination process in which, depending on at least one result of the fit determination process, one or more dimension parameters of a frame geometry of the eyeglass frame that is modified compared to the frame geometry represented by the eyeglass model data are generated as input data for a manufacturing of a real eyeglass frame controlled depending on the input data, in particular by means of a 3D printing process.

[0036] In some embodiments, the respective position of at least one of the initial head reference points is defined such that the respective initial head reference point lies on a selected body region of the head model, which was previously selected by means of algorithmic facial recognition based on the head model data or at least an image sensor-captured image of the real person's face. This enables, in particular, automation of the process without the need to define the initial head reference points during or beforehand in any other way, especially manually.

[0037] In some embodiments, at least one of the initial head reference points on the head model is defined such that it lies on one of the following body regions of the head model: eye (especially pupil), eyebrow, nose (especially nasal root or bridge or nostril), temporal bone, ear. It has been found that, firstly, these body regions can be identified very reliably and correctly using currently available facial recognition algorithms, and secondly, these body regions are of particular importance for the placement of the spectacle frame on the head because either contact points between the spectacle frame and the head typically lie in these body regions (especially in the nasal area or at the ear) or such contact points should be avoided (e.g.,(e.g., eyebrows or cheeks) or the relative placement of the actual eyeglass frame to the actual head should be oriented to such points for optometric reasons (especially the eye and pupil). The initial head reference points defined in these body regions can also be assigned to the respective body region via corresponding metadata, so that they can be selectively used for the subsequent placement steps within the fitting process. This allows for the efficient selection of the relevant initial head reference points for each placement step or subprocess from the entire set of available points.

[0038] In some embodiments, the position of at least one of the initial spectacle reference points is predefined such that the respective initial spectacle reference point lies on one of the following areas of the spectacle model: an edge (in particular a guide curve) on the lower or upper frame of a front part, a nose area, a front cheek area, or an earpiece. The edge may, in particular, be located on the side of the front part that faces the head model when the spectacle model is positioned "normally." "Normal positioning" here refers to a position in which the spectacle frame rests on the nose with its nose bridge and the earpieces rest on the respective ear root point of the head model.

[0039] In some embodiments, a set of at least three spectacle reference points is predefined in at least one area of ​​the spectacle model such that at least one of the spectacle reference points in the set is equidistant from two of the other spectacle reference points in the set. This results in an equally spaced arrangement of the spectacle reference points from the set, which can be particularly advantageous with regard to avoiding excessively large gaps and the potential inaccuracies that may arise during the fitting process when determining an optimal fit.

[0040] In some embodiments, at least a number M, with M ≥ 6, and in particular 15 ≥ M ≥ 6, of spectacle reference points are predefined on at least one of the edges, particularly guide curves, of the spectacle model. It has been found that this regularly allows for a sufficiently reliable and accurate prediction of the fit of a spectacle frame on the head for the purposes of a virtual reality trial (VTO). Furthermore, more than 15 spectacle reference points per edge generally contribute little to any significant improvement in accuracy, so that for reasons of efficiency, especially with regard to the desired real-time capability of the method given available computing resources, a limitation in this respect can also be advantageous.

[0041] In some embodiments, the head model is captured in such a way that it is represented in the head model data as a polygon mesh or set of polygon meshes, and the total number of nodes defined in this or these polygon meshes is at least 150,000. This regularly achieves a sufficiently high accuracy for VTO purposes and an effective adaptation process.

[0042] In particular, according to some embodiments, the number of initial head reference points can be chosen to be smaller, especially by one or more orders of magnitude, than the total number of nodes of the polygon mesh(es) of the head model. This is particularly advantageous with regard to improving the efficiency of the method, since it allows the number of head reference points to be considered during the placement steps to be limited without reducing the accuracy of the resolution of the head model itself (i.e., in particular, the number of its nodes). In particular, according to some embodiments, the number of initial head reference points can be limited to 50 or fewer.

[0043] In some embodiments, the spectacle model is captured in such a way that it is represented in the spectacle model data as a polygon mesh or set of polygon meshes, and the total number of nodes defined in this or these polygon meshes is at least 50,000. This regularly allows for a sufficiently high level of accuracy for VTO purposes, enabling an effective fitting process and, in particular, a graphical representation of the resulting virtual fit of the spectacle model on the head model.

[0044] In particular, according to some embodiments, the number of spectacle reference points can be chosen to be smaller, especially by one or more orders of magnitude, than the total number of nodes of the polygon mesh(es) of the spectacle model. This is also particularly advantageous with regard to increasing the efficiency of the method, since it allows the number of head reference points to be considered during the placement steps to be limited without reducing the accuracy of the resolution of the spectacle model itself (i.e., in particular, the number of its nodes). In particular, according to some embodiments, the number of spectacle reference points can be limited to 100 or fewer.

[0045] In some embodiments, texture information represented in the head model data, which represents the texture of the real person's head, is used to determine the initial head reference points. This texture can represent various prominent aspects of the person's face, such as the eyes (especially the pupils), eyebrows, ears, hair or hairline, beard, or prominent facial features such as the chin, cheekbones, nose, and mouth. Based on this texture information, particularly suitable initial head reference points for placing the spectacle model on the head model, such as the bridge of the nose, the base of the ears, the eyes, eyebrows, and nostrils, can be identified with high reliability, thus improving the placement accuracy.This can also serve to further increase efficiency, as the initial head reference points can often be identified faster based on texture information than solely on the basis of head model data representing only the head geometry, especially when the identification of the head reference points is carried out using an iterative procedure and fewer iterations are required on average when using texture information.

[0046] In some embodiments, the first placement step involves aligning a spectacle reference point located on a nose bridge of the spectacle model with a head reference point located at the bridge of the nose on the head model, and / or at least one spectacle reference point located on an earpiece of the spectacle model with a head reference point located on a temporal bone of the head model. This allows for a simple initial placement of the spectacle model on the head model.

[0047] In some embodiments, generating and outputting predictive information involves applying the overall transformation defined by combining the individual transformations determined in the preceding placement steps to the entire spectacle model in order to virtually position the entire spectacle model according to the result of the fitting process on the head model. This also serves to increase efficiency. Thus, in the individual placement steps, only one assigned transformation can initially be determined using its transformation parameters and applied to the spectacle reference points, while the spectacle model as a whole is not yet transformed.The transformation parameters of each transformation can be summarized in a corresponding transformation matrix, particularly in a manner common for translations and rotations, so that the cumulative overall transformation resulting from the individual transformations can be obtained through simple matrix algebra, especially matrix multiplication. Only this overall transformation is then applied to the entire spectacle model to position it relative to the head model based on the placement of the spectacle reference points resulting from the fitting process. Based on the results of the overall transformation, a visual representation of this placement can then be determined and output.

[0048] In some embodiments, during the adaptation process, at least one raycasting operation is performed along a virtual direction that is orthogonal to the local surface normal of the spectacle model's surface at the respective spectacle reference point from which the raycasting originates. This surface normal can, in particular, run parallel to an axis of the aforementioned local coordinate system of the spectacle model, which simplifies the calculations required for the raycasting and thus further contributes to the efficiency of the process.

[0049] In some embodiments, the number N of further placement steps iteratively performed in the fitting process is N ≥ 4, and in particular, N = 4 can be chosen. It has been shown that with N ≥ 4, very reliable process results can be achieved where the predicted fit of a real spectacle frame corresponding to the spectacle model on the person's real head (corresponding to the head model), based on the placement of the spectacle model on the head model, corresponds very well to the predicted fit. While further iteration steps beyond the fourth allow for even finer fit adjustments, they are generally not necessary to achieve a sufficiently good fit, so that N = 4 regularly represents a particularly suitable choice in terms of a compromise between efficiency and accuracy.

[0050] A second aspect of the invention relates to a system, in particular a data processing system, which has means for carrying out the method according to the first aspect, in particular according to one or more of the embodiments thereof mentioned herein.

[0051] A third aspect of the invention relates to a computer program or non-volatile computer-readable storage medium, comprising instructions which, when executed on a computer or multi-computer platform, cause the latter to execute the method according to the first aspect, in particular according to one or more of the embodiments mentioned herein.

[0052] The computer program can be stored, in particular, on a non-volatile data carrier. Preferably, this is a data carrier in the form of an optical data carrier or a flash memory module. This can be advantageous if the computer program itself is to be handled independently of a processor platform on which the one or more programs are to be executed. In another implementation, the computer program can exist as a file on a data processing unit, in particular on a server, and be downloadable via a data connection, for example, the Internet or a dedicated data connection, such as a proprietary or local network. Furthermore, the computer program can comprise a plurality of interacting individual program modules. The modules can, in particular, be configured, or at least be usable, in such a way that they function in the sense of distributed computing (i.e., distributed computing)."Distributed computing" is performed on different devices (computers or processor units that are geographically separated and connected via a data network (thus forming a multi-computer platform).

[0053] The system according to the second aspect may accordingly have a program memory in which the computer program according to the third aspect is stored. Alternatively, the system may also be configured to access an external computer program, for example on one or more servers or other data processing units, via a communication link, in particular to exchange data with it that is used during the execution of the procedure or computer program according to the first or third aspect, or that represents outputs of the computer program according to the third aspect.

[0054] The features and advantages explained in relation to the first aspect of the invention also apply accordingly to the other aspects of the invention.

[0055] Further advantages, features and applications of the present invention will become apparent from the following detailed description in conjunction with the figures.

[0056] This shows: Fig. 1 schematically the spectacle model from different perspectives to illustrate the definition of a local Cartesian coordinate system anchored to a spectacle model; Fig. 2A-C schematically a flowchart to illustrate a preferred embodiment of the method according to the invention; Fig. 3 various exemplary head reference points on a head model; Fig. 4 various exemplary spectacle reference points on a spectacle model; Fig. 5 schematically, an initial placement of the spectacle model on the bridge of the nose of the head model as part of a first placement step of procedure 200; Fig. 6 schematically, a shifting of the spectacle model from the root of the nose of the head model along the bridge of the nose based on a displacement process within the procedure from the Figuren 2A bis 2C ; Fig. 7 schematically an exemplary raycasting process within the adaptation process of the procedure from the Figuren 2A bis 2C ; Fig. 8 schematically, an exemplary adjustment step to achieve an alignment of the spectacle model with equal spacing between the left and right sides of the head model within the adjustment process of the procedure from the Figuren 2A bis 2C ; Fig. 9 schematically, an exemplary adjustment step for placing the earpieces of the spectacle model on the ear root points of the head model within the adjustment process of the procedure from the Figuren 2A bis 2C ; Fig. 10 schematically, an exemplary adjustment step for centering the nose area of ​​the spectacle model relative to the head model in a horizontal direction within the adjustment process of the procedure from the Figuren 2A bis 2C ; Fig. 11 schematically, an exemplary adaptation step to be used in the adaptation process of the procedure from the Figuren 2A bis 2C to place the glasses model on the bridge of the nose of the head model; Fig. 12 schematically, an exemplary adaptation step to be used in the adaptation process of the procedure from the Figuren 2A bis 2C to identify and, if necessary, eliminate possible collisions between the eyeglass model and the cheek areas of the head model; and Fig. 13 schematically, an exemplary adaptation step to be used in the adaptation process of the procedure from the Figuren 2A bis 2C to identify and, if necessary, eliminate possible collisions between the spectacle model and the forehead and / or bridge of the nose areas of the head model, or to adjust the spectacle model to an optimal distance from the forehead and / or bridge of the nose areas.

[0057] The same reference numerals are used throughout the figures for the same or corresponding elements of the invention.

[0058] In Fig. 1 Figure 1 illustrates an exemplary definition of a local Cartesian coordinate system 100, which is anchored at an origin 0 / 0 / 0 on the nose bridge of a pair of glasses 115. The pair of glasses 115 is shown in subfigure 115. Fig. 1 (a) already placed on a head model 105, whereby this placement can correspond in particular to an initial placement of the spectacle model 115 according to a normal spectacle positioning from the first placement step on the head model 105. In subfigure Fig. 1 (b)The X-direction of coordinate system 100 is illustrated. With reference to the spectacle model 115, it runs between the centers of the two frames of the spectacle model for the lenses. In the following, as illustrated, it is assumed that the X-direction thus runs parallel to the direction of a connecting line through the two pupils of the head model. In subfigure Fig. 1 (c) The Z-direction of coordinate system 100 is illustrated. With reference to the eyeglass model 115, it runs perpendicular to the X-direction and is furthermore aligned so that it coincides with the straight-ahead viewing direction of the head model 105. As shown in the sub-figures Fig. 1 (d) und (e) As illustrated, the Y-direction is perpendicular to the X-direction on the one hand and to the Z-direction on the other, with the axes of the coordinate system 100 coinciding with each of the three directions being aligned such that the X-axis as the first axis, the Y-axis as the second axis and the Z-axis as the third axis together with the origin 0 / 0 / 0 define the coordinate system 100 as a right-handed Cartesian coordinate system.

[0059] The three flowcharts connected via connectors "A" and "B" from the Figuren 2A , 2B und 3C Figures 200 together illustrate an exemplary embodiment of a method according to the invention for the virtual prediction of a real seat of a real eyeglass frame specific frame geometry on the head of a person with individual head geometry.

[0060] Procedure 200 is subdivided into several procedural sections 205, 230, 250, 300 and 320.

[0061] The in Fig. 2A The illustrated first procedural step 205 serves, on the one hand, to acquire data that represent the models required in the further course of the procedure, namely the real eyeglass frame and the person's head, or more precisely, the head geometry. On the other hand, it serves for the initial positioning ("rough positioning") of the eyeglass frame model on the head model.

[0062] With regard to obtaining the aforementioned data, head model data is collected in step 210, representing a model of the person's head or head geometry, specifically as a polygon model. This model is subsequently used in accordance with Fig. 1 referred to as "head model" 105. The head model data also represents a set of initial head reference points 110 predefined on the head model 105, where these points are located at positions on the head model 105 that are particularly relevant for the virtual placement of the spectacle model 115 relative to the head model 105. This is explained in more detail in the following example. Fig. 3 This is illustrated, whereby, due to their subsequent spatial proximity to the spectacle model 115, those head reference points 110 located at the eyes, eyebrows, ears, and nose of the head model are of particular relevance. In particular, one of the head reference points 110a may be located at the bridge of the nose of the head model 105, especially centered there (so-called "bridge of the nose point"). However, other helpful head reference points 110 may also be located at the mouth, jaw, forehead, or temporal bone of the head model 105.

[0063] In the next step, 215, spectacle model data is recorded, representing a model of the spectacle frame, specifically as a polygon model. This model is then compared with Fig. 1 referred to as "spectacle model" 115. The spectacle model data also represent a set of spectacle reference points 120 predefined on the spectacle model 115, which are points located at positions on the spectacle model 115 that are particularly relevant for the virtual placement of the spectacle model 115 relative to the head model 105. As in Fig. 4 As illustrated, the spectacle reference points 120 can be arranged in groups.

[0064] A first group 120a of spectacle reference points 120 can, for example, lie on a guide curve of spectacle model 115 defined by an upper rear frame edge, which, when spectacle model 115 is placed on head model 105, faces the head model 105. The guide curve can, in particular, correspond to a polygonal path of spectacle model 115 defining the edge, if the latter is defined as a polygonal model.

[0065] A second group 120b of spectacle reference points 120 can, for example, be located on a lower edge, in particular a lower guide curve, of the spectacle model 115 in an area intended to rest on the nose of the head model 105 or at least partially encompass it. This area is hereinafter also referred to as the "nose area". This lower guide curve can in turn correspond to a polygon of the spectacle model 115 that defines this edge in the nose area.

[0066] A third group 120c of spectacle reference points 120 can, for example, be located on each of the two earpieces of the spectacle model 115, particularly near their attachment to a front part of the spectacle model which has the two spectacle lens frames.

[0067] Within each of groups 120a to 120c, the spectacle reference points of the respective group 120a to 120c, or of a respective subset thereof, can be equally spaced. To achieve good accuracy for the placement of the spectacle model 115 on the head model within the framework of procedure 200, it is advantageous if at least each of groups 120a and 120b has at least six spectacle reference points. In particular, a number between 6 and 15 spectacle reference points for each of these two groups, and optionally also for group 120c, is advantageous with regard to both sufficient accuracy and efficient procedure execution.

[0068] The acquisition of head model data and / or eyeglass model data can be accomplished in various ways. On the one hand, this data may already exist before the start of procedure 200, so that in steps 210 and 215 it only needs to be made available for procedure 200, which can be done in particular by reading it from a memory or receiving a data stream via a data connection, for example from a server. On the other hand, however, it may also be the case that the head model data or the eyeglass model data are only generated within the framework of procedure 200 itself. This can be done in particular by means of a 3D scan of the person's actual head or the actual eyeglass frame. Such 3D scanning methods are known to those skilled in the art; in particular, a technique used in the authentication or unlocking of smartphones, especially more recent versions of the iPhone® (e.g., iPhone 13) from Apple, Inc., can also be used.is known. Insofar as the method 200 is carried out by a data processing device, this device can therefore be configured to receive corresponding output data from a 3D scanning device and possibly also to output a signal beforehand with which the 3D scanning device can be instructed to perform the respective 3D scan.

[0069] Furthermore, the acquisition of head model data or spectacle model data can include the generation of metadata relating to the head model or spectacle model. Such metadata can, in particular, as mentioned previously, relate to the definition of initial head reference points 110 or spectacle reference points 120. Segmentation methods can be used for this purpose, for example, with regard to the head model, methods that automatically identify different body regions on the head model, such as the eyes, mouth, nose, etc., and, based on this, define head reference points, particularly in certain selected or predetermined body regions. This can be based, in particular, on a segment-wise analysis of a 3D mesh geometry of the head model, for which splines are defined, derivations for curvatures, etc.Initial head reference points can be determined to identify particularly suitable initial head reference points, for example, at extrema, inflection points, or saddle points on the surface of the head model. It is also conceivable to determine such initial head reference points based on an image texture (especially a photo texture) and transfer them from the 2D image texture to the three-dimensional head model using a transformation. Consequently, the initial head reference points do not necessarily have to be uniformly distributed across the head model. Rather, it is advantageous to provide a higher density of initial head reference points in sections of the head model that are typically particularly relevant for the fit of a spectacle frame on the wearer's head, such as the ear and nose areas, than in less relevant areas, such as the lower cheek area.

[0070] In principle, the same approach is also possible with regard to the spectacle model. However, it is also possible there to manually define suitable spectacle reference points 120 during the design of the spectacle frame and add them as metadata to the spectacle model data.

[0071] Method 200 is based in particular on collision detection with regard to possible collisions between the eyeglass model 115 and the head model 105, and on eliminating or at least minimizing such collisions by virtually adjusting the placement of the eyeglass model 115 on the head model 105. Unlike known collision detection methods, such as those known from the field of video game programming, where simple geometric primitives anchored to a 3D model, such as virtual boxes, spheres, or cylinders, are regularly used as "hit boxes" and the collision of these primitives is detected, method 200 relies solely on the use of reference points, i.e., head reference points on the one hand and the eyeglass reference points 120 on the other, and their positioning relative to each other.The spectacle reference points 120 are ideally strategically positioned primarily so that they occur at the points on the spectacle model 115 where potential collisions with the head model 105 (and accordingly between the modeled real spectacle frame and the real head of the person) can occur or are at least particularly likely.

[0072] Now, referring again to Fig. 2A and in particular process section 205 thereof, this includes a further step 220 in which a spectacle reference point 120, in particular a spectacle reference point 120b, which is located on the nose bridge of the spectacle model, and in particular the origin 0 / 0 / 0 according to Fig. 1 can be placed on a corresponding initial head reference point 110 on the nose of the head model 105, in particular on the nasal root point 110a.

[0073] To establish an initial orientation of the spectacle model 115 based on this placement, a step 225 is also provided within process section 205, in which the spectacle model 115 is aligned, in particular rotated about its local X-axis and its local Y-axis, such that at least one, preferably each, of the earpieces of the spectacle model 115 is aligned on its respective side of the head model with a corresponding selected initial head reference point 110b or 110c on the temporal bone, in particular passing through or over it. Steps 220 and 225 for the initial (rough) placement and (rough) alignment, which constitutes a first placement step of process 200, are additionally described in Fig. 5 illustrated.

[0074] The translational and rotational transformations resulting from steps 220 and 225 can now be summarized, in particular in a corresponding first transformation matrix M 0, which represents the initial placement of the eyeglass model 115 on the head model 105 resulting from these transformations, parameterized and stored in a memory 125 for later use.

[0075] In a second, also in Fig. 2A In the illustrated, optional process section 230, also referred to here as the "translation process," a step 235 first checks whether, according to a user input 240 at a user interface (GUI), a virtual translation of the spectacle model 110 on the head model 105 is desired by the user. If this is the case (235 - yes), a virtual translation of the spectacle model 110 corresponding to the user input 240 takes place in a translation direction 250a running parallel to the bridge of the nose of the head model 105, in particular in the direction of a head reference point 110d on the bridge of the nose in the region of the tip of the nose. Otherwise (235 - no), step 245 is omitted. Step 245 is described in more detail in Fig. 6 illustrated. The translational transformation that may result from step 245 can now be parameterized and stored in a corresponding second transformation matrix M 1, which represents this transformation, and temporarily stored in a memory, in particular in memory 125, for later use.

[0076] As a Fig. 2B The illustrated third procedure step 250 of procedure 200 now involves an iterative fine-tuning of the placement of the spectacle model 115 on the head model 105. The third procedure step 250 is an exemplary execution of the aforementioned adjustment process, and each iteration corresponds to an exemplary further placement step of this adjustment process. The third procedure step, or adjustment process 250, is based in particular on the use of raycasting, which is now initially performed with reference to Fig. 7 This will be explained in more detail before the individual steps of the adaptation process 250 are discussed in detail.

[0077] After the execution of the first process step 205 and, if applicable, the second process step 230, the rough positioning and orientation of the spectacle model 115 relative to the head model 115 are known. This allows certain assumptions to be made, such as the assumption that the spectacle reference points 120b of the nose area of ​​the spectacle model 115 are located near the nose area of ​​the head model 105. This makes it possible to define additional head reference points 140, which are added to the head reference points 110 already defined in the head model data. Accordingly, these additional head reference points 140 are also, or even primarily, used in the adaptation process 250, so that relevant aspects of the geometry of the head model 105 can be included in greater detail without having to use the entire head model 105 with its much higher complexity.Instead, the fitting process continues to rely solely on head reference points 110 and 140 on the one hand, and the spectacle reference points 120 on the other. This ensures the desired high efficiency while simultaneously achieving high fitting accuracy when placing the spectacle model 115 on the head model 105.

[0078] In calculating the respective transformation(s) in the individual steps 260 to 285 of the fitting process 250, each spectacle reference point 120 included in the respective step is assigned one or more corresponding head reference points, in particular additional head reference points 140. The additional head reference points 140 are determined as needed from the respective spectacle reference point 120 using raycasting. Raycasting is a well-established and widely used method that is already available as a tool in many interactive 3D environments.

[0079] In Fig. 7 One such raycasting process is illustrated by way of example. Two raycastings are always performed from a spectacle reference point 120: firstly, using a first virtual ray 130a in a defined direction, and secondly, using a second virtual ray 130b in the opposite direction, where these directions can each be described by a corresponding direction vector.

[0080] The initial placement of the spectacle model 115 on the head model 105, defined by means of the rough positioning in the first process step 205, determines the directions of the (virtual) rays 130a and 130b such that they point towards the head model (see ray 130a) and away from it (see ray 130b).

[0081] In some of steps 260 to 285, one or more of the spectacle reference points 120 can each serve as a starting point for a plurality of raycastings in different directions.

[0082] Fig. 7 This shows, as an example, a step of fine positioning within the adaptation process 250, in which the aim is to ensure that the nose area of ​​the spectacle model 125 rests on the nostrils of the nose of the head model 105.

[0083] For each of a set of spectacle reference points 120b in the nasal region of the spectacle model 115, two local direction vectors for rays 130a and 130b are defined in this process step. In this example, these local direction vectors point, in particular, along the Y-axis (ray 130b) and in the opposite direction (ray 130b). Performing two raycastings for each spectacle reference point 120 in the set is necessary because a spectacle reference point 120 can initially lie either inside or outside the volume defined by the surface of the head model 105. The respective relative position of each spectacle reference point 120 with respect to the head model 105 (inside, outside, or on the head surface) can be determined by performing two raycastings. Depending on which ray hits the surface of the head model 105, the spectacle reference point 120 lies either inside or outside the head model.

[0084] For example, if a spectacle reference point 120b, considered here as an example, is located below the surface of the head model that models the nose of the head model, i.e., within the nose geometry, a collision is detected, and the spectacle reference points 120, and subsequently the entire spectacle model based on them, must be transformed to resolve this collision. Conversely, if all spectacle reference points 120 are located outside the nose area of ​​the head model, then the spectacle reference points 120, and consequently the spectacle model 115 as a whole, must be transformed in a direction towards the nose area so that the spectacle model 115 rests on the nostrils of the head model 105 within the nose area.

[0085] At point 135, where a raycasting beam, in this example beam 130a, intersects the head model 105 ( Fig. 7 (a)) an additional head reference point 140 of the head model 105 is generated ad hoc, in particular temporarily ( Fig. 7 (b) For the current step of the adaptation process 250, the spectacle reference point 120b, used as the starting point for raycasting, is assigned to the newly determined additional head reference point 140 for the purpose of calculating the transformation for this current step. After the current step is completed and the associated transformation has been determined and stored, the previously determined additional head reference point 140 can be discarded and the process can continue with the next step.

[0086] Now, with reference to… Fig. 2B and the Figuren 8 bis 12 The individual steps of the adaptation process 250 are explained in detail.

[0087] To control the iteration provided for in the adaptation process 250, an iteration counter i can first be initialized in step 255, for example with the value "1". This is followed by an iteration loop that is executed repeatedly until a predetermined termination criterion is reached. In the present example, as shown in steps 290 and 295, the termination criterion is the reaching of a number N (here, for example, N=4) of iterations, which is also specified in step 255. Alternatively, a termination criterion could also be defined that depends on the degree of adaptation achieved between the fit of the spectacle model 115 and the head model 105.

[0088] As part of the iteration in the adaptation process 250, steps 260 to 295 are executed sequentially in each iteration run.

[0089] In step 260, which is described in more detail in Fig. 8 As illustrated, the spectacle model 115 is rotated around the local Y-axis (perpendicular to the image plane through a point in the center of the nose bridge of the spectacle model 115, which may coincide with the origin 0 / 0 / 0, as illustrated here) in order to achieve an alignment of the spectacle model 115 that is equidistant from the left and right sides of the head model 105.

[0090] To determine the rotation angle, raycasting is performed from the respective spectacle reference points 120c and 120d on the two earpieces of spectacle model 115 to the temple area of ​​head model 105. This results in points of impact for the rays, represented as thick arrows, on head model 105, at which additional head reference points 140a and 140b are newly defined. Existing head reference points 110e and 110f on the two ears of head model 105 can also be used. With the help of the newly found additional head reference points 140a and 140b at the temples and the existing head reference points 110e and 110f, the rotation angle can now be determined with the goal of ensuring that the earpieces, i.e., more precisely, those of the spectacle model, are equidistant from the ears and temples on both the left and right sides of the head model.More precisely, this means that, firstly, the spectacle reference points 120c and 120d on the two earpieces are the same distance from their respective head reference points 110e and 110f on the ears, and secondly, the spectacle reference points 120c and 120d are the same distance from the two additional head reference points 140a and 140b on the temples, each in a direction pointing away from the respective head reference point on the head model.

[0091] As from Fig. 8 As can be seen, this adaptation goal has not yet been achieved at the beginning of step 260. Instead, the spectacle reference point 120c lies inside the head model 105 (negative distance), while the spectacle reference point 120d lies outside the head model 105 (negative distance). After execution of step 260, however, a symmetrical situation should have been established in this respect, in which the two spectacle reference points 120c and 120d lie either outside or on the head model (or its surface) and, in this respect, have the same distance along the beam direction of the respective raycasting towards the head model 105.

[0092] Furthermore, in Fig. 9 In more detail illustrated, step 265 of the adaptation process 250 involves a rotational transformation depending on the relative position of a head reference point 110 on an ear of the head model 105, which may coincide in particular with the ear root point 110e, to a corresponding spectacle reference point 120 on an earpiece of the spectacle model 115 associated with the ear, in order to achieve as an adaptation goal a placement of the spectacle model on the head model in which the earpiece rests on the ear root of the ear.

[0093] This step 265 can also rely solely on initial head reference points 120. These initial head reference points 120 can include, in particular, head reference points on the left and right sides of the head model, especially the ear root points 110e and 110f. Additionally, a spectacle reference point on the nose bridge of the spectacle model, which can again coincide with the origin 0 / 0 / 0 of the local coordinate system, is used. The spectacle model 115 is then rotated around the local X-axis and / or the local Z-axis with the spectacle reference point on the nose bridge as the pivot point, so that the earpieces on the left and right sides of the head model and spectacle model each rest on the corresponding left and right head model points (especially the ear root points 110e and 110f) on the corresponding ear of the head model 105.

[0094] Furthermore, in Fig. 10 As illustrated in more detail, step 270 of the fitting process 250 involves centering the nose area of ​​the spectacle model 115 in the horizontal direction, i.e., along the X-direction, relative to the nose of the head model 105. For this purpose, raycastings parallel to the X-axis of the local coordinate system 100 of the spectacle model 115 are performed, starting from the spectacle reference points 120b in the nose area of ​​the spectacle model 115. At the points where the raycasting rays strike the head model 105, new additional head reference points 140 are generated in the nose area of ​​the head model 105. In the specific example of the Fig. 10 Starting from six spectacle reference points 120b, 12 raycastings are performed. In this example, a total of four additional head reference points 140 are temporarily identified on the bridge of the nose. The raycastings to the two upper reference points, however, do not result in any further additional reference points 140, because no hits occur on the head model 105.

[0095] The respective distances between the spectacle reference points 120b on the one hand and the corresponding, newly found head reference points 140 in the nasal area of ​​head model 105 on the other are used to standardize the distances between the spectacle model 115 and the nose in the nasal area of ​​head model 105 on the left and right sides of the nose. The transformation required for this is a rotation about the Y-axis. While in Fig. 10 While only spectacle reference points 120b in the nose area are visualized on the anterior lower guide curve of the spectacle model, for the most accurate possible centering of the spectacle model 115 on the head model 105 within the framework of step 270, further spectacle reference points distributed along the direction of the Z-axis (i.e. perpendicular to the drawing plane) on the spectacle model 115, in particular on its nose pad, are also advantageous.

[0096] Furthermore, in Fig. 11 In more detail illustrated step 275 of the adjustment process 250, the position of the nose area of ​​the spectacle model 115 is adjusted in the vertical direction, i.e. along the Y-direction, relative to the nose of the head model 105.

[0097] Similar to the preceding step 270, raycasting rays are emitted from spectacle reference points 120b in the nasal region of the spectacle model, specifically from the same spectacle reference points 120b as in step 270. However, this time they are emitted in a vertical direction, since step 270 aims to align the spectacle model 115 with the bridge of the nose of the head model 105.

[0098] For this purpose, the distance of each of the spectacle reference points 120b to its corresponding additional head reference point 140 on the surface of the face of the head model 105, determined based on raycasting, is calculated, provided such a head reference point 140 exists. This information is used to determine the rotation angle of a rotation around an axis of rotation running between the ear root points 110e and 110f, such that after this transformation, the nose pad of the spectacle model rests on the nostrils of the head model 105. As in the preceding step 270, in Fig. 11 Only spectacle reference points 120b are shown on the front lower guide curve of the spectacle model in its nose area, whereas for the most accurate possible positioning of the spectacle model 115 on the head model 105 in step 275, further spectacle reference points distributed along the direction of the Z-axis (i.e. perpendicular to the drawing plane) on the spectacle model 115, in particular on its nose pad, may also be advantageous.

[0099] In the specific example of the Fig. 11 Additional head reference points were found in the nasal area of ​​the head model for all raycastings. The spectacle model 115 must now be rotated towards the nose during the transformation so that at least one of the spectacle reference points 120b touches its corresponding additional head reference point 140 or its immediate vicinity on the nose of the head model 115.

[0100] In another, in Fig. 12 In step 280 of the fitting process 250, which is illustrated in more detail, a check is performed to determine whether there are collisions between the cheek regions of the head model 105 and the lower frame area of ​​the spectacle model 115. If necessary, a transformation is carried out to eliminate such collisions. In this transformation, the spectacle model 115 is rotated around an axis of rotation passing through the head reference points 110e and 110f located at the base of the ears, so that such collisions in the cheek area are eliminated. However, if such collisions are still detected in the last iteration N of the fitting process 250, and the spectacle model subsequently rests on the cheek of the head model due to the final execution of step 280, an output can optionally be initiated at a human-machine interface to report this.

[0101] In the specific example of the Fig. 12 Two additional head reference points are identified in the cheek area of ​​head model 105. This confirms that there is no collision between the spectacle model 115 and the cheek area. Furthermore, the distance between the spectacle model 115 and the cheek area is sufficient that no transformation needs to be performed in step 280 in this case.

[0102] In the Fig. 13 In the final adjustment step 280 of each iteration of the adjustment process 250, illustrated in more detail, a raycasting check is again performed to determine whether there are collisions between spectacle reference points 120a on the upper frame area of ​​the spectacle model 115 and the forehead or bridge of the nose of the head model. If this is the case (i.e., if at least one of the spectacle reference points 120a lies on or within the head model, as indicated by the raycasting), then such collisions are resolved by a translational transformation of the spectacle reference points in a direction pointing away from the forehead along a specific temple vector. The temple vector can be defined, in particular, such that it corresponds to the projection of a line between the endpoints of an earpiece of the spectacle model onto the local Y / Z plane.

[0103] Even if no such collisions are detected, such a translation can be performed to adjust the optimal distance of the spectacle model from the forehead. Unless otherwise specified in the translation process 230, it can be assumed for the translation that the real spectacle frame corresponding to the spectacle model 115 should ideally be worn close to the forehead or the bridge of the nose. Accordingly, the spectacle reference points 120 are transformed during this translation so that this corresponds to setting a predefined optimal distance of the spectacle model 115 to the forehead or bridge of the nose. In the specific example of the Fig. 13 Although no collisions exist, there is a significant distance between the upper frame of the spectacle model and the bridge of the nose of head model 105. Therefore, the spectacle model must be transformed translationally somewhat in the direction of the bridge of the nose.

[0104] The individual transformations defined within the framework of the adaptation process 250 per iteration can now be summarized, in particular in a corresponding further transformation matrix M i+1 per iteration i, which represents these transformations of the i-th iteration, parameterized and stored in a memory, in particular again in the memory 125, for later use within the framework of the procedure 200.

[0105] With reference to Fig. 2CA third process step 300 of process 200 is now described, which concerns the generation and output of predictive information for predicting the actual fit of the real spectacle frame corresponding to the spectacle model on the person's real head. The third process step 300 builds on the results of the preceding process steps, in particular the fitting process 250, in which fine positioning of the spectacle model 115 on the head model 115 was carried out for the purpose of fit optimization.

[0106] First, in step 305 of the third process section 300, the entire spectacle model 115 is transformed, i.e., in the case of a polygon model, all nodes of the polygon model. This transformation results from the transformation matrices M 0 to M N+1, determined in the preceding process sections and stored in memory 125, which are now read from memory 125. This transformation is for the individual transformations described above, in particular through appropriate sequential execution or matrix multiplication.

[0107] Here, an advantage of method 200 regarding its efficiency comes into play. While the individual transformations described so far in the preceding process steps were only applied to the set of spectacle reference points 120—that is, to a typically very limited number of points, which is sensibly chosen to be much smaller than the number of nodes in the polygon model of the spectacles (spectacle model)—these individual transformations could be executed very quickly, especially in real time. Only now, in step 305, i.e., after the transformation parameters have been determined, is the resulting overall transformation applied to the entire spectacle model. Thus, this more computationally intensive step only needs to be performed once.

[0108] The transformed spectacle model 115 can now be output in a suitable manner in step 315 using data representing it, in particular graphically on a user interface, in order to provide a user of the procedure 200 with an optical impression of the fit of the spectacle model 115 on the head model 105 and thus a reliable prediction of the actual fit of the real spectacle frame corresponding to the spectacle model 115 on the real head of the person corresponding to the head model 105.

[0109] Optionally, a fit determination process 310 may also be provided, which serves in particular to make a particularly quantified prediction regarding the expected fit and the expected wearing comfort of the real spectacle frame on the real head of the person on the basis of information collected within the framework of procedure 200.

[0110] In particular, within the framework of the fit determination process 310, (i) a respective ear temple length on the spectacle model 115 between the attachment point of the ear temple on the front part of the spectacle model 115 and the respective corresponding ear root point can be calculated and (ii) the respective contact pressure of each ear temple of the real spectacle frame on the assigned side of the person's real head can be estimated on the basis of the calculated ear temple length, the relative position of the ear temple of the spectacle model 115 to the head model 105 and information from which a spring force of the respective ear temple can be deduced at a given deflection from a rest position (e.g. geometry and material data).

[0111] It is also possible, based on the repositioned eyeglass model 115 as part of the adjustment process, to determine certain proportions of eyeglass model 115 and head model 105 that are meaningful with regard to the fit, such as the position of the upper eyeglass frame of eyeglass model 115 relative to the eyebrows on head model 105.

[0112] Furthermore, during the fitting process, certain pieces of information determined during the fitting process can be used to infer aspects of the fit of the spectacle model to the head model 105 or of the actual spectacle frame to the person's actual head. This information can include, in particular, specific distances determined during raycasting or distances resulting after the respective transformations of the fitting steps have been carried out. For example, how many of the spectacle reference points 120b in the nose area of ​​the spectacle model 115 lie on the nose area of ​​the head model 105 after completion of the fitting process 250. Alternatively, or additionally, relative spatial positions, particularly based on proportions, between spectacle reference points and head reference points can be considered.For example, the position of head reference points associated with the pupils can be considered in relation to reference points on the upper and lower edges of the spectacle frame, from which a viewing point for the respective eye can be determined. According to another example, the relative spatial position of certain spectacle reference points—head reference points associated with the temples or certain (other) contour points of the face—can be considered to determine whether the spectacle frame width appears optically harmonious in relation to the width of the face.

[0113] In the case of a static standard pair of glasses, i.e., not custom-made glasses, an output can also be generated, particularly at a user interface, which provides information regarding the specific fit, in particular whether the real frame corresponding to the glasses model 115 will fit or whether it would be better to try on or choose a different frame (especially again using VTO) in order to achieve an improved fit.

[0114] In contrast, particularly in the case of custom-made spectacles, a fourth process step 320 may be provided in process 200, in which, firstly, in a step 325, suitable dimensional parameters for the spectacle frame of the custom-made spectacles are determined based on the fit information obtained in the third process step 300, and subsequently, in a further step 330, the additive manufacturing of a custom-made spectacle frame based on these dimensional parameters is triggered, controlled, or carried out, particularly depending on the corresponding capabilities of a system used to carry out process 200. In particular, the dimensional parameters can be determined, for example, such that the aforementioned contact pressure of the earpieces on the head lies within a predetermined pressure range that is typically considered appropriate and comfortable for the individual or generally for a "standard spectacle wearer" defined as a reference.The length of the earpieces and any bending point within them can also be precisely determined, so that any bent part of the earpieces runs directly behind the ear when the eyeglass frame is correctly positioned on the person's head.

[0115] While at least one exemplary embodiment has been described above, it should be noted that a large number of variations exist. It should also be noted that the described exemplary embodiments are merely non-limiting examples, and it is not intended to restrict the scope, applicability, or configuration of the devices and methods described herein. Rather, the preceding description will provide the person skilled in the art with guidance for implementing at least one exemplary embodiment. It is understood that various modifications to the function and arrangement of the elements described in an exemplary embodiment can be made without deviating from the subject matter defined in the appended claims. REFERENCE MARK LIST

[0116] 100 Coordinate system 105 Head model 110 Head reference points 110a Head reference point at the bridge of the nose of the head model (nasal root point) 110b Head reference point on the right temporal bone of the head model 110c Head reference point on the left temporal bone of the head model 110d Head reference point on the bridge of the nose of the head model 110e, f Head reference points on the ears of the head model, in particular ear root points 115 Eyeglass model 120 Eyeglass reference points 120a Eyeglass reference points on an upper guideline of the eyeglass model 120b Eyeglass reference points on a guideline in the nasal area of ​​the eyeglass model 120c, b Eyeglass reference points on an earpiece of the eyeglass model 125 Memory 130a First virtual ray in raycasting, extending in the first direction 130b Second virtual ray in raycasting, extending in the opposite direction to the first direction 135Point of impact of a virtual raycasting beam on a surface (e.g.Node, edge or polygon) of the head model 140 additional (ad-hoc) head reference point, determined by raycasting 140a-bs special additional (ad-hoc) head reference points, determined by raycasting 200 Method according to an embodiment of the invention 200-330 Method sections or steps of the method 200.

Claims

1. A method (200) for virtually predicting a real seat of a real eyeglasses frame of a specific frame geometry on the head of a person with an individual head geometry, the method (200) comprising: obtaining head model data representing a three-dimensional virtual head model (105) of at least a part of the head of the person; obtaining eyeglasses model data representing a three-dimensional virtual eyeglasses model (115) of at least a part of the frame geometry of the eyeglasses frame; a first placement step in which a point set of a plurality of eyeglasses reference points on the eyeglasses model (115) predefined by or on the basis of the eyeglasses model data is aligned relative to a point set of one or more initial head reference points on the head model (105) predefined by or on the basis of the head model data in order to define an initial virtual placement of the eyeglasses model (115) on the head model (105); an adaptation process with one or more further placement steps carried out iteratively following the first placement step, in which one or more additional head reference points (140) on the head model (105) are determined in each case by means of ray casting on the basis of a respective virtual beam emanating from one or more of the eyeglasses reference points (120) and are used to determine at least one parameter value of a parameterized translational and / or rotational transformation of the eyeglasses reference points (120) for adapting the virtual placement of the eyeglasses reference points (120) relative to the head model and the correspondingly defined transformation is carried out with respect to the eyeglasses reference points (120); and generating and outputting prediction information which represents a prediction of the real seat of the real eyeglasses frame on the head of the person on the basis of the virtual placement of the eyeglasses reference points (120) relative to the head reference points resulting within the scope of the iteration from the transformation of the most recently carried out placement step.

2. The method (200) according to claim 1, wherein: the determination of the transformation for the respective current placement step comprises an adaptation of at least one parameter of the transformation depending on the relative position of at least one of the eyeglasses reference points (120) to at least one of the additional head reference points (140) determined in the current placement step in order to achieve an adaptation goal defined with respect to this position.

3. The method (200) according to claim 2, wherein the first placement step or at least one of the placement steps of the adaptation process further comprises: determining in each case at least one additional head reference point (140) on the head model (105) per head side by means of ray casting on the basis of a virtual beam emanating from a respective eyeglasses reference point (120) on an ear temple of the eyeglasses model corresponding to the respective same head side; and adapting at least one parameter of the transformation depending on the respective relative position of each of these eyeglasses reference points (120) to the additional head reference point determined emanating therefrom in order to achieve an alignment of the eyeglasses model (115) equidistant with respect to the left and the right head side of the head model (105) as an adaptation goal.

4. The method (200) according to one of the preceding claims, wherein at least one of the placement steps of the adaptation process for determining the transformation for this placement step comprises one or more of the following sub-processes, in particular according to the following order: (a) adapting at least one parameter of the transformation depending on the relative position of a head reference point on an ear of the head model to a corresponding eyeglasses reference point (120) of the eyeglasses model in order to achieve, as an adaptation goal, a placement of the eyeglasses model (115) on the head model (105) in which the ear temple rests on the ear root of the ear; (b) determining in each case at least one additional head reference point on the nose of the head model (105) by means of ray casting in a direction running at least partially parallel to the anatomical horizontal axis of the head model (105) on the basis of a virtual beam emanating from a respective eyeglasses reference point (120) on a nose region of the eyeglasses model configured to at least partially enclose the nasal lobes of a nose of the head model (105); and adapting at least one parameter of the transformation depending on the respective relative position of each of these eyeglasses reference points (120) to the additional head reference point determined emanating therefrom in order to achieve, as an adaptation goal, a centering of the nose region of the eyeglasses model (115) on the head model (105) with respect to the direction running parallel to the anatomical horizontal axis; (c) determining in each case at least one additional head reference point on the nose of the head model (105) by means of ray casting in a direction running at least partially parallel to the anatomical longitudinal axis of the head model (105) on the basis of a virtual beam emanating from a respective eyeglasses reference point (120) on a nose region of the eyeglasses model configured to at least partially enclose the nasal lobes of a nose of the head model (105); and adapting at least one parameter of the transformation depending on the respective relative position of each of these eyeglasses reference points (120) to the additional head reference point determined emanating therefrom in order to achieve, as an adaptation goal, an at least pointwise support of the nose region of the eyeglasses model (115) on the nasal lobes of the nose of the head model (105); (d) checking whether, according to the current placement of the eyeglasses model (115) on the head model (105), there is a collision of the eyeglasses model with a cheek region of the head model (105); and if this is the case, adapting at least one parameter of the transformation in order to achieve, as an adaptation goal, a cancellation of the collision with a cheek region; (e) checking whether, according to the current placement of the eyeglasses model (115) on the head model (105), there is a collision of the eyeglasses model with a forehead or nose bridge region of the head model (105); and if this is the case, adapting at least one parameter of the transformation in order to achieve, as an adaptation goal, a cancellation of the collision with the forehead or nose bridge region.

5. The method (200) according to one of the preceding claims, further comprising a displacement process carried out between the first placement step and the adaptation process, in which the eyeglasses reference points (120) are subjected to a translational transformation in a direction along or parallel to a nose bridge line of the head model (105), which extends between a head reference point located at the nose root of the head model (105) and a head reference point located on the nose bridge of the head model (105).

6. The method (200) according to one of the preceding claims, further comprising a fit determination process following the adaptation process, which comprises at least one of the following sub-processes: - determining, on the basis of the placement of the eyeglasses model (115) on the head model (105) resulting from the adaptation process and for at least one ear temple of the eyeglasses model (115), a respective length of an ear temple portion, which extends between a front part of the eyeglasses model (115) and the support point of the ear temple on an ear root point on the head model; - estimating a contact pressure of an ear temple of the real eyeglasses frame corresponding to the ear temple of the head model (105) on the real head of the person by means of calculating a virtual contact pressure of the ear temple of the eyeglasses model (115) on the head model (105) as a function of the determined length of the ear temple portion and known elasticity properties of the real ear temple; - determining, on the basis of the placement of the eyeglasses model (115) on the head model (105) resulting from the adaptation process and for at least one sub-region of the eyeglasses model (115) comprising a plurality of eyeglasses reference points (120), which portion of these eyeglasses reference points (120) rests on the head model (105); - determining, on the basis of the placement of the eyeglasses model (115) on the head model resulting from the adaptation process, for at least one selected eyeglasses reference point (120), the respective relative spatial position thereof with respect to at least one point which corresponds to a head reference point or is determined depending on the respective positions of a plurality of head reference points (110); - estimating the quality of a fit of the real eyeglasses frame on the real head of the person depending on the estimated contact pressure, the determined portion of the eyeglasses reference points (120) of the sub-region resting on the head model (105), and / or on one or more of the determined relative spatial positions; - determining an evaluation of the fit on the basis of a comparison of the estimated quality of the fit with reference data corresponding thereto; - outputting the determined quality and / or evaluation of the fit as a component of the prediction information, wherein the method further preferably comprises a measurement parameter determination process in which, depending on at least one result of the fit determination process, one or more measurement parameters of a frame geometry of the eyeglasses frame modified with respect to the frame geometry represented by the eyeglasses model data are generated as input data for a production of a real eyeglasses frame controlled depending on the input data.

7. The method (200) according to one of the preceding claims, wherein the respective position of at least one of the initial head reference points (110) is defined such that the respective initial head reference point lies at a selected body region of the head model (105) which was previously selected by means of algorithmic face recognition on the basis of the head model data or at least one image of the face of the real person obtained by image sensor.

8. The method (200) according to one of the preceding claims, wherein at least one of the initial head reference points (110) on the head model (105) is defined such that it lies at one of the following body regions of the head model (105): eye, eyebrow, nose, temporal bone, ear.

9. The method (200) according to one of the preceding claims, wherein the respective position of at least one of the initial eyeglasses reference points (120) is or is predefined such that the respective initial eyeglasses reference point (120) lies at one of the following regions of the eyeglasses model (115): an edge on the lower or upper frame of a front part, a nose region, a front cheek region, an ear temple, wherein in at least one of the regions of the eyeglasses model (115) in each case a set of at least three eyeglasses reference points is or is preferably predefined such that at least one of the eyeglasses reference points (120) of the set has the same distance with respect to two of the other eyeglasses reference points (120) of the set, wherein at least in each case six eyeglasses reference points (120) are or are preferably predefined at at least one of the edges of the eyeglasses model (115).

10. The method (200) according to one of the preceding claims, wherein the head model (105) is obtained such that it is represented in the head model data as a polygon mesh or set of polygon meshes and the total number of the nodes defined in this or these polygon meshes is at least 150,000, wherein the number of the initial head reference points (110) is preferably selected to be less than the total number of the nodes of the polygon mesh or mesh of the head model (105).

11. The method (200) according to one of the preceding claims, wherein the eyeglasses model (115) is obtained such that it is represented in the eyeglasses model data as a polygon mesh or set of polygon meshes and the total number of the nodes defined in this or these polygon meshes is at least 50,000, wherein the number of the eyeglasses reference points (120) is preferably selected to be less than the total number of the nodes of the polygon mesh or mesh of the eyeglasses model (115).

12. The method (200) according to one of the preceding claims, wherein texture information represented in the head model data which represents a texture of the head of the real person is used to determine the initial head reference points (110).

13. The method (200) according to one of the preceding claims, wherein in the first placement step a eyeglasses reference point which lies on a bridge of the nose of the eyeglasses model (115) is brought to coincide with a head reference point which lies at the nose root of the head model (105) and / or at least one eyeglasses reference point which lies on an ear temple of the eyeglasses model (115) is brought to coincide with a head reference point which lies at a temporal bone of the head model (105).

14. The method (200) according to one of the preceding claims, wherein the generating and outputting of prediction information comprises an application of the overall transformation defined by the individual transformations defined in the preceding placement steps to the entire eyeglasses model in order to virtually place the eyeglasses model (115) overall according to the result of the adaptation process on the head model (105).

15. The method (200) according to one of the preceding claims, wherein in the adaptation process at least during a ray casting this ray casting is carried out along a virtual direction which runs orthogonally to the local surface normal of the surface of the eyeglasses model (115) at the respective eyeglasses reference point (120) from which the ray casting emanates.

16. A device, in particular a data processing device, which comprises means for carrying out the method according to one of the preceding claims.

17. A computer program or a non-transitory computer-readable storage medium, comprising instructions which, when they are executed on a computer or a multi-computer platform, cause said computer or platform to carry out the method (200) according to one of claims 1 to 15.