Stereoscopic display methods and devices allowing ophthalmic defects to be taken into account

Abstract

The present disclosure relates to a method (20) of display of stereoscopic images to a user, by a display device (10) comprising a processing unit (11) and a display unit (12), said display unit comprising a first display area for displaying a first image for one eye of the user and a second display area for displaying a second image for the other eye of the user, said method comprising: - obtaining (S20) a user-specific size-related correction representative of a difference in size perception between the eyes of the user, - estimating (S21) a direction of vision of the eyes of the user and estimating, depending on the estimated direction of vision, a central point targeted by the eyes of the user on the display unit, - modifying (S22) the first image and / or second image by applying the size-related correction with respect to the estimated central point, - displaying (S23) the first image and second image, obtained after modification, on the first display area and second display area, respectively.

Description

Description Title: Stereoscopic display methods and devices for addressing ophthalmic defects technical field

[0001] The present invention belongs to the field of stereoscopic display devices, and relates more particularly to methods and stereoscopic display devices enabling users with ophthalmic defects to perceive stereoscopic images in relief.

[0002] The present invention finds a particularly advantageous, although by no means limiting, application in the case of stereoscopic display devices embedded in motor vehicles. State of the art

[0003] Stereoscopic display devices, as is well known, display stereoscopic images that allow a user to see a scene in three dimensions. Stereoscopic images consist of a pair of images of the same scene, representing that scene from slightly different viewpoints that roughly correspond to the positions of a user's eyes relative to the scene. Thus, a stereoscopic display device has a first display area and a second display area, each showing an image intended for the user's left eye and a second image intended for the user's right eye, respectively.

[0004] There are different types of stereoscopic display devices. Some stereoscopic display devices require special glasses to allow the user to perceive the scene in relief, while other display devices, called autostereoscopic, allow the user to perceive the scene in relief without the need for special glasses or other accessories.

[0005] Some users with refractive errors are unable to perceive depth when viewing stereoscopic images on a stereoscopic display device. For example, some users may have a difference in size perception between their left and right eyes (aniseikonia), such as due to anisometropia or, for a user who wears glasses, due to a significant difference in thickness between the left and right lenses. Such a difference in size perception between a user's left and right eyes impairs their stereoscopic vision, preventing them from perceiving depth when viewing stereoscopic images. Other refractive errors, such as strabismus (crossed eyes), can also affect stereoscopic vision. Description of the invention

[0006] The present invention aims to overcome all or part of the limitations of prior art solutions, in particular those set out above, by proposing stereoscopic display methods and devices to improve the stereoscopic vision of users with certain ophthalmic defects, in particular a difference in size perception between the right and left eye (aniseikonia).

[0007] To this end, a method for displaying stereoscopic images to a user is proposed according to a first aspect, by a display device comprising a processing unit and a display unit, said display unit comprising a first display area of ​​a first image for one eye of the user and a second display area of ​​a second image for the other eye of the user, said method comprising: obtaining a relative size correction, specific to the user, representative of a difference in size perception between the eyes of said user, an estimation of a vision direction of the user's eyes and an estimation, as a function of the estimated vision direction, of a central point aimed by the user's eyes on the display unit, a modification of the first image and / or the second image by applying the relative size correction with respect to the estimated central point,a display of the first image and the second image, obtained after modification, on the first display area and the second display area, respectively.

[0008] Thus, the proposed solution modifies stereoscopic images by applying a relative size correction between the first image (for example, intended for the left eye) and the second image (for example, intended for the right eye). This relative size correction aims to compensate for the difference in size perception between the user's eyes. One image will therefore be enlarged (or reduced) relative to the other image in the stereoscopic image pair. This enlargement (or reduction) is further applied relative to a central point targeted by the user on the display unit of the display device, so that the image obtained after enlargement (or reduction) varies with the targeted central point. Indeed, the inventors have found that taking into account the central point targeted by the user when applying the relative size correction improves the restoration of stereoscopic vision.

[0009] In particular modes of implementation, the process according to the first aspect may further include one or more of the following optional characteristics, taken individually or in all technically possible combinations.

[0010] In particular modes of implementation of the process according to the first aspect, the relative size correction comprises two correction components to be applied along two orthogonal axes.

[0011] In particular modes of implementation of the process according to the first aspect, the direction of vision of the user's eyes is estimated based on images of the user acquired by an optical sensor.

[0012] In particular modes of implementation of the process according to the first aspect, the display device being installed in a vehicle, the optical sensor is a camera of a driver monitoring system of said vehicle.

[0013] In particular modes of implementation, the process according to the first aspect includes filtering the temporal variations of the estimated vision direction and / or the estimated central point by means of a low-pass filter, with a time constant between 125 ms and 500 ms.

[0014] In particular modes of implementation of the process according to the first aspect: the process involves obtaining a relative position correction, specific to the user, representative of a difference in direction of vision between the eyes of said user, the first image and the second image, obtained after modification, are displayed on the first display area and the second display area, respectively, offset from each other by the relative position correction.

[0015] Such provisions make it possible to restore depth perception for a user who also has strabismus.

[0016] In particular modes of implementation of the process according to the first aspect, obtaining the relative size correction specific to the user by the display device involves selecting the relative size correction to be applied from among several relative size corrections specific to said user, based on an estimation of an optical state of said user, the optical state of said user comprising at least one parameter among the following parameters: the presence or absence of glasses worn by the user, a type of glasses worn by the user, the presence or absence of lenses worn by the user.

[0017] In particular modes of implementation of the process according to the first aspect, with the display device installed in a vehicle, the optical state of the user is estimated based on images acquired by at least one external or internal camera of said vehicle.

[0018] According to a second aspect, a method is proposed for determining, by a display device, a relative size correction for stereoscopic images, specific to a user, representative of a difference in size perception between the user's eyes, the display device comprising a processing unit, a user interface and a display unit, said display unit comprising a first display area of ​​a first calibration image for one eye of the user and a second display area of ​​a second calibration image for the other eye of the user, the first calibration image and the second calibration image being complementary in the absence of an ophthalmic defect, said method comprising an iteration of the following steps until a stopping criterion is met: an estimation of a user's eye-viewing direction and an estimation, based on the estimated eye-viewing direction, of a central point aimed by the user's eyes on the display unit, a modification of the first calibration image and / or the second calibration image by applying a candidate relative size correction relative to the estimated central point, a display of the first calibration image and the second calibration image, obtained after modification, on the first display area and the second display area, respectively, a receipt of user feedback via the user interface, an evaluation of the stopping criterion based on the user feedback received, in which the relative size correction is determined for the user based on one or more candidate relative size corrections validated by the user feedback.

[0019] In particular modes of implementation, the process according to the second aspect may further include one or more of the following optional characteristics, taken individually or in all technically possible combinations.

[0020] In particular modes of implementation of the process according to the second aspect, the first calibration image represents an object in a first color and the second calibration image represents the same object in a second color different from the first color so that, in the absence of an ophthalmic defect, a user perceives the object entirely in a third color which corresponds to the mixture of the first and second colors.

[0021] In particular modes of implementation of the process according to the second aspect, the first calibration image and the second calibration image are stereoscopic images representing a first object and a second object such that, in the absence of an ophthalmic defect, a user perceives the second object as being in the background with respect to the first object.

[0022] In particular modes of implementation of the process according to the second aspect, the relative size correction comprises two correction components to be applied along two orthogonal axes.

[0023] In particular modes of implementation of the process according to the second aspect, the direction of vision of the user's eyes is estimated based on images of the user acquired by an optical sensor.

[0024] In particular modes of implementation of the process according to the second aspect, the display device being installed in a vehicle, the optical sensor is a camera of a driver monitoring system of said vehicle.

[0025] In specific implementation methods, the process according to the second aspect includes filtering of temporal variations of the estimated vision direction and / or the estimated central point by means of a low-pass filter with a time constant between 125 ms and 500 ms.

[0026] In particular modes of implementation of the method according to the second aspect: the method further comprises a determination of a relative position correction, specific to the user, representative of a difference in direction of vision between the eyes of said user, the first calibration image and the second calibration image, obtained after modification, are displayed on the first display area and the second display area, respectively, offset from each other by a candidate relative position correction.

[0027] According to a third aspect, a computer program product is proposed comprising instructions which, when executed by a display device comprising a processing unit and a display unit comprising a first display area for one eye of a user and a second display area for the other eye of the user, configure said display device to implement a process according to any one of the implementation modes of this disclosure.

[0028] According to a fourth aspect, it is proposed a computer-readable recording medium on which is recorded a set of instructions which, when executed by a display device comprising a processing unit and a display unit comprising a first display area for one eye of a user and a second display area for the other eye of the user, configure said display device to implement a process according to any of the implementation modes of this disclosure.

[0029] According to a fifth aspect, a display device is proposed comprising a processing unit and a display unit comprising a first display area for one eye of a user and a second display area for the other eye of the user, configured to implement a process according to any one of the implementation modes of this disclosure.

[0030] According to a sixth aspect, a vehicle, such as a motor vehicle, is proposed, comprising a display device according to any one of the embodiments of this disclosure. Presentation of the figures

[0031] The invention will be better understood upon reading the following description, given by way of non-limiting example, and made with reference to the figures which represent: [Fig. 1] Figure 1: A schematic representation of an example of a stereoscopic display device, [Fig. 2] Figure 2: A diagram illustrating the main steps of an example of implementation work of a stereoscopic image display process, [Fig. 3] Figure 3: a representation of examples of image modifications as a function of a relative size correction and the central point targeted by the user, [Fig. 4] Figure 4: A diagram illustrating the main steps of an example of implementing a process for determining relative size correction for a user.

[0032] In these figures, identical references from one figure to another designate identical or analogous elements. For clarity, the elements shown are not to scale unless otherwise indicated.

[0033] Furthermore, the order of steps shown in these figures is given only as a non-limiting example of this disclosure which may be applied with the same steps performed in a different order and / or with steps performed in parallel and / or jointly. Description of the implementation methods

[0034] Figure 1 schematically represents an example of an embodiment of a display device 10. As illustrated by Figure 1, the display device 10 comprises a processing unit 11 and a display unit 12.

[0035] The display unit 12 is a stereoscopic display unit, meaning that it has a first display area showing a first image for one eye (e.g., the left eye) of the user and a second display area showing a second image for the other eye (e.g., the right eye) of the user. The first and second images are therefore stereoscopic in that they are intended to be presented to different eyes of the user of the display device 10.

[0036] As noted above, there are various types of stereoscopic display units, and the choice of a particular type of stereoscopic display unit for the display device 10 represents only one non-limiting implementation variant of this disclosure. Some stereoscopic display units, for example, require special glasses to allow the user to perceive the scene depicted in three dimensions.This is the case, for example, when the first and second images are displayed with lights of different polarizations, separable for the user's right and left eyes using polarized glasses, or when the first and second images are displayed successively (time-multiplexed, the first and second display areas potentially overlapping) and separable for the user's right and left eyes using blocking glasses (synchronized with the display, which alternately block the user's left and right eyes), etc. Other display units, called autostereoscopic, allow the user to perceive the scene in 3D without the need for special glasses or other accessories. This is the case, for example, when the first and second images are spatially multiplexed on the same screen and separable. the user's right and left eyes by means of, for example, parallax barriers or a lenticular system, or on two screens worn by the user respectively in front of the user's right eye and in front of the user's left eye (for example in the case of virtual or augmented reality glasses), etc.

[0037] In the case where the display device 10 is installed in a vehicle, such as a motor vehicle, the display unit 12 is, for example, a remote autostereoscopic display unit located away from the user (i.e., not worn by the user). For example, the display unit 12 corresponds to the dashboard of the motor vehicle. However, nothing precludes considering other types of stereoscopic display units and / or other operating environments besides a vehicle, as illustrated in other examples.

[0038] The processing unit 11 includes, for example, one or more processors (CPU, DSP, GPU, FPGA, ASIC, etc.) and one or more memories (magnetic hard drive, electronic memory, optical disk, etc.) in which, for example, a computer program product is stored, in the form of a set of program code instructions to be executed by the processor(s) to implement all or part of the process steps that will be described below.

[0039] In some embodiments, and as illustrated in Figure 1, the display device 10 may optionally include a user interface 13, allowing the user to interact with the display device 10. Generally, any type of user interface can be considered, and the choice of a particular type for the user interface 13 represents only a non-limiting variant of the implementation of this disclosure. For example, the user interface 13 may include one or more of the following: a touchscreen (in which case the user interface 13 may be fully or partially integrated into the display unit 12), a scroll wheel, a joystick, a keyboard, a mouse, etc. The user interface 13 may, for example, be a smartphone communicating with the processing unit 11.

[0040] Figure 2 schematically represents the main steps of an example of the implementation of a method 20 for displaying stereoscopic images implemented by the display device 10.

[0041] As illustrated by Figure 2, the display method 20 includes a step S20 of obtaining a relative size correction, specific to the user, representative of a difference in size perception between the eyes of said user.

[0042] As previously stated, relative size correction aims to compensate for the difference in size perception between the eyes of the user of the display device 10. Thus, after applying relative size correction, an object in a scene represented by stereoscopic images is perceived as being essentially the same size by both eyes of the user in question; that is, the difference in size perception is essentially canceled out by applying relative size correction. Relative size correction is therefore "specific" to the user in question in that it is adapted to substantially cancel out the difference in size perception of that user (the difference in size perception, and therefore the associated relative size correction, varying from one user to another).

[0043] Relative size correction can be applied as either an enlargement or reduction factor to one of the stereoscopic images—that is, to either the first or the second image. Alternatively, relative size correction can be applied as an enlargement factor to one of the stereoscopic images (for example, the first image) and a reduction factor to the other (for example, the second image if the enlargement factor is applied to the first). Thus, depending on the example, applying relative size correction means modifying either one of the stereoscopic images (the first or the second image) or both stereoscopic images (both the first and the second image).

[0044] Furthermore, a magnification or reduction factor can consist of a single component applied to the entire image, or it can comprise several components, for example, two components applied along two orthogonal axes, as is the case for a user with significant astigmatism. Indeed, in the case of astigmatism, the correction provided by a lens is not the same in all directions and is broken down into two distinct corrections along two orthogonal axes, which do not necessarily correspond to the horizontal and vertical axes. In such cases, the relative size correction can have two different components applied along these two orthogonal axes, which aim to compensate for the differences in corrections provided by the lenses.These two different axes can be predefined (e.g. horizontal and vertical) or their definitions in the display unit's coordinate system 12 can be included in the relative size correction, especially if the axes do not correspond to the horizontal and vertical axes and / or are user-specific.

[0045] The S20 step of obtaining the user-specific relative size correction consists, for example, of receiving said relative size correction via the user interface 13 or from equipment separate from the display device 10, for example, equipment worn by the user (smartphone, connected glasses, etc.) or a remote server in which relative size corrections for one or more users are stored.

[0046] Following another example, the user-specific size-based relative correction can be calculated during the S20 retrieval step from information received via the user interface 13 or from equipment separate from the display device 10, for example, equipment worn by the user (smartphone, smart glasses, etc.) or a remote server in which information relating to one or more users is stored. This information corresponds, for example, to the data from the user's medical prescription. For example, the magnification (or reduction) factor of relative size correction can be calculated based on the number K of diopters of difference between the user's left and right eyes. For example, relative size correction can enlarge an image by 1.5% multiplied by K, which corresponds to a magnification factor of (100 + 1.5x) / 100.

[0047] Following another example, the S20 acquisition step can implement a method for determining the relative size correction, aimed at calibrating the difference in size perception between the user's eyes by presenting them with calibration (stereoscopic) images. A non-limiting example of a method 40 for determining the relative size correction is described below with reference to Figure 4.

[0048] It should be noted that, in some examples, it is also possible to have several specific relative size corrections for the same user. For example, the relative size correction to be applied to a user may vary depending on the user's optical condition, i.e., whether or not they wear glasses, and / or the type of glasses worn in the case of a user with multiple pairs of glasses, and / or the presence or absence of contact lenses. In such a case, the S20 acquisition step may also, in some examples, involve selecting the relative size correction to be applied from among several relative size corrections specific to that user, based on an estimate of the user's optical condition (presence or absence of glasses, type of glasses worn, presence or absence of contact lenses, etc.).For example, external equipment on a motor vehicle, such as surround-view cameras, can be used to assess the user's vision, including detecting the presence or absence of eyeglasses and, potentially, the type of eyeglasses worn. Internal equipment, such as the camera in a driver monitoring system (DMS), can also be used, either alternatively or in addition, to assess the user's vision. A DMS camera, for instance, can assess the user's vision by comparing the apparent size of the two eyes and can even detect the combination of eyeglasses and contact lenses (typically monofocal lenses to correct aniseikonia and glasses to correct presbyopia).This detection is carried out, for example, by comparing the apparent size of the eyes.

[0049] As illustrated in Figure 2, the display method 20 includes a step S21 of estimating a direction of vision of the user's eyes and estimating, as a function of the estimated direction of vision, a central point aimed by the user's eyes on the display unit 12.

[0050] Indeed, as previously indicated, relative size correction aims to compensate for the difference in size perception between the eyes of the user of the display device 10. Thus, after applying relative size correction, an object in a scene represented by Stereoscopic images are perceived as having essentially the same size (and the same shape in the case of different components applied along orthogonal axes) for both eyes of the user. However, the inventors found that correcting the difference in perceived size was not always sufficient to achieve good superposition and restoration of depth perception. Indeed, if the enlargement (or reduction) is performed relative to an arbitrary point such as the center or a corner of the image, this will generally be insufficient to restore depth perception. Instead, this enlargement (or reduction) should be performed relative to a central point that corresponds to a point the user is looking at on the display unit 12. This central point is estimated during step S21 by estimating the user's viewing direction relative to said display unit 12.Thus, the way in which the enlargement (or reduction) is carried out depends on the central point targeted by the user.

[0051] In general, any method known to a person skilled in the art can be implemented to estimate the user's vision direction relative to the display unit 12 and, from the estimated vision direction, to estimate the center point aimed at by the user on the display unit 12. The choice of a particular estimation method is only a non-limiting variant of implementation of this disclosure.

[0052] For example, the user's line of sight can be estimated based on images of the user acquired by one or more optical sensors, which may be an integral part of the display unit 10 or may belong to other equipment communicating with the display unit 10. If the display unit 10 is installed in a motor vehicle, the optical sensor might be, for example, a camera in a driver monitoring system (DMS) of that motor vehicle. The line of sight can thus be estimated in a coordinate system of that driver monitoring system and conventionally transferred to a coordinate system of the display unit 12 to determine the central point aimed at by the user.

[0053] In practice, the user's line of sight must be tracked over time to follow changes in the central point targeted by the user on the display unit 12. This is because the user does not always look in the same place, and it is therefore necessary to track shifts in their line of sight. In such a case, it can be advantageous to implement a low-pass filter to filter the temporal variations of the estimated line of sight and / or the estimated central point, in order to avoid tracking erratic movements of the line of sight and / or the central point. For example, the low-pass filter, optionally implemented to filter these temporal variations, has a time constant between 125 ms and 500 ms.

[0054] As illustrated in Figure 2, the display method 20 also includes a step S22 of modifying the first image and / or the second image by applying the Relative size correction with respect to the estimated central point.

[0055] As previously mentioned, relative size correction can be applied as a factor of enlargement (or reduction) to one of the stereoscopic images, that is, to either the first or the second image. Alternatively, relative size correction can be applied as a factor (enlargement or reduction) to one of the stereoscopic images (for example, the first image) and a factor (enlargement or reduction) to the other stereoscopic image (for example, the second image). Thus, depending on the example, applying relative size correction means modifying only one of the stereoscopic images (the first or the second image), or modifying both stereoscopic images (both the first and the second image).

[0056] The modification applied to one of the stereoscopic images during step S22 corresponds to an enlargement (or reduction) relative to the estimated center point targeted by the user. This is, for example, a scaling operation where the center corresponds to the estimated center point and the ratio corresponds to the enlargement (or reduction) factor. This modification may also conventionally involve resampling the digital pixels of the image being modified. Indeed, the various digital pixels of the stereoscopic images are initially positioned at the physical pixel level of the display unit 12. Due to the enlargement or reduction, these digital pixels no longer correspond to the positions of the physical pixels of the display unit 12, and it is therefore necessary to resample the modified image to obtain new digital pixels for the grid of physical pixels of the display unit 12.In general, any resampling method can be implemented, and the choice of a particular method is only one non-limiting variant of the implementation of this disclosure. For example, resampling may involve bilinear or bicubic interpolation.

[0057] As illustrated in Figure 2, the display method 20 includes a step S23 of displaying the first image and the second image, obtained after modification, on the first display area and the second display area of ​​the display unit 12, respectively.

[0058] Figure 3 schematically represents examples of modifications applied to one of the stereoscopic images. In these examples, it is considered, without limitation, that the modification applied corresponds to an enlargement applied to the first image, which must be displayed by the first display area of ​​display unit 12.

[0059] As illustrated by part a) of Figure 3, the first image represents a scene containing a triangular object 30. Parts b) and c) of Figure 3 represent the object 30 obtained after enlarging the first image obtained by considering different central points aimed by the user on the display unit 12 (first display area), but in considering the same magnification factor in both cases.

[0060] In the example illustrated by part b) of figure 3, the central point 31a targeted by the user is located in the lower left corner of the first display area of ​​the display unit 12. The object obtained after enlarging the first image, considering the central point 31a as the center of the homothety, is designated by the reference 30a.

[0061] In the example illustrated by part c) of figure 3, the central point 31b aimed at by the user is located substantially at the center of the object 30, that is to say that the user is looking at this object 30. The object obtained after enlarging the first image, taking into account the central point 31b as the center of the homothety, is designated by the reference 30b.

[0062] As can be seen in parts b) and c) of Figure 3, the object 30a, 30b displayed on the first display area of ​​the display unit 12 varies with the central point 31a, 31b aimed by the user on said display unit 12. It is this consideration of the central point actually aimed by the user (rather than considering an arbitrary point independent of the user's direction of vision) that makes it possible to modify the stereoscopic images in such a way as to improve the perception of relief of the object represented, by the user.

[0063] As illustrated by Figure 2, in some implementation modes, steps S21, S22, S23 can be iterated, for example to track the movement of the center point aimed at by the user on the display unit 12 and / or to process new stereoscopic images (i.e., process a new first image / second image pair), for example in the case of successive stereoscopic images from a stereoscopic (3D) video stream.

[0064] As mentioned above, other ophthalmic defects can affect a user's depth perception. For example, a user with strabismus will have significantly different vision directions for their right and left eyes, so the central point they are looking at is not the same for both eyes. Without correction, the user's right and left eyes will not be looking at the same point in the scene depicted in the first and second images, which can impair depth perception if there is a significant discrepancy.

[0065] To restore depth perception in cases of strabismus, the display method 20 may, in certain implementation examples, include a step (not shown in the figures) for obtaining a relative position correction, specific to the user, representing a difference in the direction of vision between the user's eyes. In these examples, during step S23, the first and second images obtained after modification are then displayed on the first and second display areas, respectively, offset from each other by the relative position correction.

[0066] The relative position correction aims to compensate for the difference between the central points targeted by the user's right and left eyes on display unit 12. Thus, after applying the relative position correction, the first and second images are positioned offset from each other so that the right and left eyes of The user, although presenting different directions of vision, is looking at substantially the same place in the scene represented by said first image and said second image.

[0067] The step of obtaining the user-specific relative position correction consists, for example, of receiving said relative position correction via the user interface 13 or from equipment separate from the display device 10, for example, equipment worn by the user (smartphone, smart glasses, etc.) or a remote server in which relative position corrections for one or more users are stored.

[0068] As another example, the user-specific relative position correction can be calculated during the retrieval step from information received via the user interface 13 or from equipment separate from the display device 10, for example, equipment worn by the user (smartphone, smart glasses, etc.) or a remote server in which information relating to one or more users is stored. This information corresponds, for example, to the data from a user's medical prescription.

[0069] Following another example, the user-specific relative position correction can be determined by separately estimating, during step S21, the user's right eye vision direction and left eye vision direction.

[0070] As another example, the acquisition step may implement a method for determining the relative position correction, aimed at calibrating the difference in size perception between the user's eyes by presenting them with calibration (stereoscopic) images. A non-limiting example for determining the relative position correction is described below.

[0071] Figure 4 schematically represents the main steps of an example of the implementation of a method 40 for determining the relative size correction of stereoscopic images.

[0072] The determination process 40 is, for example, implemented by a display device 10 as shown in Figure 1. It should be noted that it is possible to use the same display device 10 to implement both the display process 20 and the determination process 40 (the determination process 40 being implemented, for example, during step S20 of obtaining the display process 20). It is also possible, following other examples, to use for the determination process 40 a display device that is physically separate from the display device 10 used for the display process 20; the display device 10 used for the display process 20 being, for example, mounted in a motor vehicle, but not the one used for the determination process 40.

[0073] In principle, the determination method 40 relies on displaying a first calibration image on the first display area of ​​the display unit 12, and a second calibration image on the second display area. The first and second calibration images are complementary in the absence of an ophthalmic defect. By "complementary in the absence of an ophthalmic defect," it is meant that when the first and second calibration images are displayed by the display unit 12 Without applying any correction, these images form a predetermined pattern recognizable to a user without an ophthalmic defect. For a user with a difference in size perception between their eyes, when the first and second calibration images are displayed by the display unit 12 without applying any relative size correction, they do not form this predetermined pattern. For a user with a difference in size perception between their eyes, the predetermined pattern is perceived only when the first and second calibration images are displayed by the display unit 12 with a relative size correction applied to compensate for that user's difference in size perception.Therefore, the determination method 40 will test several candidate relative size corrections to modify the first and second calibration images until the user recognizes the predetermined pattern. This means that the candidate relative size correction, which led the user to recognize the predetermined pattern, is well-suited to the user. For example, it is possible to search for the candidate relative size correction that yields the best results in terms of complementarity between the modified first and second calibration images. In this case, the relative size correction determined specifically for this user corresponds to the candidate relative size correction validated by the user as providing the best results.As another example, it is possible to search for several candidate relative size corrections for which the user can recognize the predetermined pattern. In this case, the specific relative size correction for that user can be determined from several candidate relative size corrections validated by the user, allowing them to recognize the predetermined pattern.

[0074] In general, different types of complementary calibration images can be considered, provided that the first and second calibration images allow the user to detect when the difference in perceived size between their eyes is correctly corrected by a relative candidate size correction, through the recognition of said predetermined pattern. The choice of a particular type of calibration image is only one non-limiting implementation variant of this disclosure.

[0075] For example, the first calibration image might depict an object in one color, and the second calibration image might depict the same object, in the same position, in a second color different from the first. The first and second colors could be primary colors. Thus, in the absence of an ophthalmic defect, or when the difference in size perception is correctly corrected, the user perceives the object entirely in a third color, which corresponds to a mixture of the first and second colors. The object entirely in the third color therefore corresponds to the predetermined pattern to be detected. For example, the object corresponds to a shape Simple geometric shapes (square, triangle, circle, etc.) filled with either the first color (first calibration image) or the second color (second calibration image). In some examples, the first and second colors may be adapted to compensate for potential color vision deficiencies in the user (color blindness, etc.), thus allowing a user with color vision deficiency to detect the predetermined pattern (an object entirely in the third color).

[0076] Following another example, the first and second calibration images can be stereoscopic images representing a first and second object such that, in the absence of an ophthalmic defect, or when the difference in size perception is correctly corrected, the user perceives the second object as being in the background relative to the first object. Having the first object in the foreground relative to the second object thus corresponds to the predetermined pattern to be detected. For example, the first and second calibration images can represent several identical objects (e.g., squares, triangles, circles, etc.), only one of which (the first object) should be perceived as being in the foreground relative to the other objects (the second objects).

[0077] Following another example, the first calibration image might represent a first object, and the second calibration image might represent a second object that differs from the first but has a complementary shape. Thus, in the absence of an ophthalmic defect, or when the difference in size perception is correctly corrected, the user perceives a third object with a recognizable shape that corresponds to a combination of the shapes of the first object and the third object. For example, the first and second objects might correspond to complementary semicircles (or semi-disks), so that the third object corresponds to a complete circle (or disk), and so on.

[0078] As illustrated in Figure 4, the determination method 40 includes a step S40 of estimating the direction of vision of the user's eyes and an estimation, based on the estimated direction of vision, of a central point aimed by the user's eyes on the display unit 12. Everything previously described with reference to the estimation step S21 of the display method 20 is also applicable to the estimation step S40 of the determination method 40.

[0079] As illustrated in Figure 4, the determination method 40 also includes a step S41 of modifying the first calibration image and / or the second calibration image by applying a candidate relative size correction with respect to the estimated center point. As previously stated, the determination method 40 will test several candidate relative size corrections until the user recognizes the predetermined pattern for the first and second calibration images. For example, the initially considered candidate relative size correction is zero, meaning no relative size correction is applied. Following another example, the correction The relative candidate size initially considered may be a relative correction of theoretical size calculated from a user's prescription, in which case the determination process 40 aims to refine this relative correction of theoretical size (to take into account, in particular, the possible natural evolution of the ophthalmic defect). Everything described previously with reference to step S22 of modifying the display process 20 is also applicable to step S41 of modifying the determination process 40.

[0080] As illustrated in Figure 4, the determination method 40 also includes a step S42 for displaying the first and second calibration images, obtained after modification, on the first and second display areas of the display unit 12, respectively. Everything described previously with reference to the display step S23 of the display method 20 also applies to the display step S42 of the determination method 40.

[0081] As illustrated in Figure 4, the determination process 40 includes a step S43 of receiving user feedback via the user interface 13 of the display device 10, and a step S44 of evaluating a predetermined stopping criterion based on the user feedback received.

[0082] User feedback determines whether the stopping criterion is met. Generally, if user feedback indicates that other candidate relative size corrections should be considered (for example, if the user feedback modifies the considered candidate relative size correction), then the stopping criterion is not met. Conversely, if user feedback indicates that it is no longer necessary to consider other candidate relative size corrections (after the user feedback has previously validated at least one candidate relative size correction as detecting the predetermined pattern), then the stopping criterion is met. For example, if the user does not validate the candidate relative size correction via user interface 13 (if they have not detected the predetermined pattern), the user feedback may indicate that the candidate relative size correction is not suitable and / or needs to be modified.The modification required to update the candidate relative size correction is, for example, included in the user feedback. When the user feedback modifies the candidate relative size correction, this means that other candidate relative size corrections must be tested, so the stopping criterion is not checked (reference S44a in Figure 4), and steps S40, S41, S42, S43, and S44 are iterated. The modification may, for example, change a magnification or reduction factor along one or two axes, or change the axes along which the different components of the candidate relative size correction are considered (for example, by rotating the previously considered orthogonal axes with a predetermined angular step, for example, 15°), etc.The relative candidate size correction obtained after modification is then used during the next iteration of steps S40, S41, S42, S43, S44.

[0083] The user can validate the relative candidate size correction used when they have detected The predetermined pattern occurs when the first and second calibration images, obtained after modification and displayed by the display unit 12, are perceived as complementary by the user. Generally, the stopping criterion can be verified, depending on the examples considered, as soon as a candidate size relative correction is validated by the user, or after a plurality of candidate size relative corrections have been validated by the user. For example, it is possible to iterate steps S40, S41, S42, S43, S44 until a predetermined number Ne (Ne greater than or equal to 1) of candidate size relative corrections is validated, or until the user considers that a sufficient number of candidate size relative corrections have been validated, and so on.When user feedback indicates that it is not necessary to test other candidate relative size corrections (for example, if user feedback validates the candidate relative size correction used and the number of candidate relative size corrections validated by the user is equal to Ne, or if user feedback explicitly indicates that it is no longer necessary to test other candidate relative size corrections and at least one candidate relative size correction has been previously validated by the user, etc.), the stopping criterion is checked (reference S44b in Figure 4).

[0084] As illustrated in Figure 4, the process 40 includes a step S45 for determining the user-specific relative size correction to be used by the display process 20, based on the candidate relative size corrections validated by the user. If the user has validated only one candidate relative size correction, the relative size correction determined specifically for that user corresponds to the candidate relative size correction validated by the user via the user interface 13. If the user has validated several candidate relative size corrections, then the relative size correction determined specifically for that user is based on all or some of the candidate relative size corrections validated by the user.For example, the relative size correction determined specifically for this user corresponds to the average, possibly weighted, of the candidate relative size corrections validated by the user via user interface 13.

[0085] As indicated above, other ophthalmic defects can affect a user's depth perception. For example, a user with strabismus will have significantly different vision directions for their right and left eyes, so the central point they are aiming at is not the same for both eyes. In such cases, it may be necessary to determine a relative positional correction for this user, in addition to the relative size correction. To this end, the determination method 40 may, in certain implementation examples, include a step (not shown in the figures) for determining a user-specific relative positional correction that represents a difference in vision direction between the user's eyes.As with relative size correction, it is possible to test different candidate relative position corrections, by displaying the first calibration image and the second calibration image, obtained afterwards. The modification, on the first and second display areas respectively, is offset from each other by a relative candidate position correction. If applicable, the stopping criterion is not checked if the user feedback modifies the relative candidate position correction. As with the relative size correction, the user can validate one or more relative candidate position corrections, and the relative position correction specifically determined by that user is based on the relative candidate position correction(s) validated by that user.

[0086] More generally, it should be noted that the implementation and realization methods considered above have been described as non-limiting examples, and that other variants are therefore conceivable.

[0087] Specifically, the method 40 for determining the relative size correction was described by considering that the central point targeted by the user on the display unit 12 was estimated from an estimated viewing direction during step S40, and that this estimated central point was used during the modification step S41. However, following other examples, there is nothing to preclude not performing the estimation step S40 and instead making the modification relative to a predetermined point on the display unit 12, which corresponds, for example, to the center of an object represented by the first and second calibration images, which the user is instructed to view during the determination method 40. In other words, the estimation step S40 does not need to be performed if the first and second calibration images are such that the central point targeted by the user is known a priori.

Claims

Demands 1. A method (20) for displaying stereoscopic images to a user, by means of a display device (10) comprising a processing unit (11) and a display unit (12), said display unit comprising a first display area for a first image for one eye of the user and a second display area for a second image for the other eye of the user, said method comprising: obtaining (S20) a relative size correction, specific to the user, representative of a difference in size perception between the eyes of said user, estimating (S21) a direction of vision of the user's eyes and estimating, as a function of the estimated direction of vision, a central point aimed by the user's eyes on the display unit, modifying (S22) the first image and / or the second image by applying the relative size correction with respect to the estimated central point,a display (S23) of the first image and the second image, obtained after modification, on the first display area and the second display area, respectively.

2. Method (20) according to claim 1, wherein the relative size correction comprises two correction components to be applied along two orthogonal axes.

3. Method (20) according to any one of the preceding claims, wherein the direction of vision of the user's eyes is estimated based on images of the user acquired by an optical sensor.

4. Method (20) according to claim 3, wherein, the display device being installed in a vehicle, the optical sensor is a camera of a driver monitoring system of said vehicle.

5. Method (20) according to any one of the preceding claims, comprising filtering the temporal variations of the estimated vision direction and / or the estimated central point by means of a low-pass filter, with a time constant between 125 ms and 500 ms.

6. A method (20) according to any one of the preceding claims, wherein obtaining (S20) the user-specific relative size correction by the display device involves selecting the relative size correction to be applied from among several user-specific relative size corrections, based on an estimation of an optical state of said user, the optical state of said user comprising at least one of the following parameters: the presence or absence of glasses worn by the user, a type of glasses worn by the user, the presence or absence of lenses worn by the user.

7. Method (20) according to claim 6, wherein the display device is installed In a vehicle, the user's optical condition is estimated based on images acquired by at least one external or internal camera of said vehicle.

8. A method (40) for determining, by a display device (10), a relative size correction for stereoscopic images, specific to a user, representative of a difference in size perception between the user's eyes, the display device comprising a processing unit (11), a user interface (13) and a display unit (12), said display unit comprising a first display area of ​​a first calibration image for one eye of the user and a second display area of ​​a second calibration image for the other eye of the user, the first calibration image and the second calibration image being complementary in the absence of an ophthalmic defect, said method comprising an iteration of the following steps until a stopping criterion is met: an estimation (S40) of a viewing direction of the user's eyes and an estimation, as a function of the estimated viewing direction,from a central point targeted by the user's eyes on the display unit, a modification (S41) of the first calibration image and / or the second calibration image by applying a candidate relative size correction with respect to the estimated central point, a display (S42) of the first calibration image and the second calibration image, obtained after modification, on the first display area and the second display area, respectively, a reception (S43) of user feedback via the user interface, an evaluation (S44) of the stopping criterion based on the user feedback received, in which the relative size correction is determined based on one or more candidate relative size corrections validated by the user feedback.

9. Method (40) according to claim 8, wherein the first calibration image represents an object in a first color and the second calibration image represents the same object in a second color different from the first color so that, in the absence of an ophthalmic defect, a user perceives the object entirely in a third color which corresponds to the mixture of the first color and the second color.

10. Method (40) according to claim 8, wherein the first calibration image and the second calibration image are stereoscopic images representing a first object and a second object such that, in the absence of an ophthalmic defect, a user perceives the second object as being in the background with respect to the first object.

11. Method (40) according to any one of claims 8 to 10, wherein the relative size correction comprises two correction components to be applied along two orthogonal axes.

12. A method (40) according to any one of claims 8 to 11, comprising a filtering temporal variations of the estimated vision direction and / or the estimated central point using a low-pass filter with a time constant between 125 ms and 500 ms.

13. Product computer program comprising instructions which, when executed by a display device (10) comprising a processing unit (11) and a display unit (12) comprising a first display area for one eye of a user and a second display area for the other eye of the user, configure said display device to implement a method (20, 40) according to any one of the preceding claims.

14. Display device (10) comprising a processing unit (11) and a display unit (12) comprising a first display area for one eye of a user and a second display area for the other eye of the user, configured to implement a method (20, 40) according to any one of claims 1 to 12.

15. Vehicle comprising a display device (10) according to claim 14.

Images