Device and method for viewing a part of the face of a user

Abstract

A device (1) for viewing a part of the face of a user (U), having at least one screen (2) and a camera (3) for continuously acquiring images of the face of a user (U) so as to allow the user to view images displayed by the screen when they are facing the camera (2), the acquired images comprising a region of interest (Iroi) of the face of the user (U), the device (1) being configured to generate, on the basis of the acquired images, enlarged stabilized images of the region of interest (Iroi) and to display them on the screen (2), an image processing being carried out prior to display of the images, the images displayed on the screen (2) being processed so as to at least partially compensate for at least one vision defect of the user (U).

Description

[0001] Description

[0002] Title: Device and method for viewing a part of the face of a user

[0003] Technical field

[0004] The present invention relates to the carrying out of a cosmetic treatment, in particular to the application of a make-up or therapy product, and more particularly aims to improve the visual comfort and application performance of persons suffering from a vision defect during this cosmetic treatment.

[0005] Prior art

[0006] It is common in bathrooms to find magnifying mirrors, usually supported by adjustable brackets, that allow people to observe a region of the face close up. These mirrors have been known for a very long time. However, using them is still difficult because the user must position themselves precisely in relation to the mirror. In addition, the image is very easily distorted.

[0007] More recently, make-up assistance devices have been proposed that use a smartphone equipped with a camera capable of capturing images of the user’s face in the front position. After acquisition, these images are enlarged and displayed on the screen, with the option of adjusting the brightness if necessary.

[0008] The patent US 9,594,948 B2 discloses a virtual make-up device that allows a user to continuously acquire images of their face and to view different virtually applied make-up on a screen.

[0009] The application CN 109063671 discloses a make-up assistance device configured to acquire a series of images of a user’s face and to display altered images containing information intended to guide the user when making up.

[0010] The application CN111914604A also discloses a device for virtually viewing a colouring of the eyebrows.

[0011] In addition, it is common for individuals to develop presbyopia after a certain age, or to have myopia or hyperopia, which may require them to wear glasses. On the one hand, wearing glasses can make making up difficult, especially in the region of the eye, by hindering the individual when they are retained. On the other hand, the user may encounter difficulties in carrying out precise and exact motions when applying make-up without glasses.

[0012] Consequently, there is still a need to improve the devices for assisting with the application of a cosmetic product in order to overcome the aforementioned problems.

[0013] Summary of the invention

[0014] The invention aims to meet this need and, according to one of its aspects, achieves this by means of a device for viewing a part of the face of a user, having at least one screen and a camera for continuously acquiring images of the face of the user so as to allow the user to view images displayed by the screen when they are facing the camera, the acquired images comprising a region of interest of the face of the user, the device being configured to generate, on the basis of the acquired images, enlarged stabilized images of the region of interest and to display them on the screen, an image processing being carried out prior to display of the images, the images displayed on the screen being processed so as to at least partially compensate for at least one vision defect of the user, in particular, a defect chosen from presbyopia and hyperopia.

[0015] Courtesy of the invention, the user's visual comfort is improved while making up their face.

[0016] In particular, the user can apply a cosmetic product without their glasses or corrective lenses. The product can thus be applied relatively precisely and quickly. In particular, the invention can make it possible, if the aforementioned image processing is carried out, for people suffering from presbyopia, myopia or hyperopia to make up without wearing glasses or corrective lenses.

[0017] Enlarging images allows the user to view details in the region being observed more easily than in the absence of enlargement.

[0018] The term “stabilized image” should be understood to mean that the displayed image is centred in relation to the same landmark or reference point, even when the camera is subjected to certain vibrations or small-magnitude movements of the user. Stabilizing displayed images can provide a more precise view. In particular, the user is more likely to benefit from the viewed images, even if they have a disease that affects their motor abilities, for example Parkinson’s disease, multiple sclerosis or essential tremor. Blurring due to certain movements of the user and / or the camera can be at least partially avoided.

[0019] Furthermore, the device according to the invention is relatively simple to use and allows the user to use it autonomously and ergonomically.

[0020] The term “camera” here encompasses all digital image acquisition devices operating in the visible or near infrared, of any resolution and technology.

[0021] The term "enlarged image" should be understood to mean that the image is displayed in a larger size than that acquired, the displayed image being related to at least one region of the acquired image by a predefined enlargement (also called "zoom") factor. Preferably, zooming is carried out at the same location for each of the acquired images. For example, zooming affects a region located substantially in the centre of each of the acquired images, or at one of the corners of each of the acquired images.

[0022] The resolution of an image refers to the number of pixels contained in the image per unit length.

[0023] The definition of an image refers to the number of pixels that an image contains in width and height, i.e. the number of columns and the number of rows of pixels.

[0024] The term “predetermined region of interest” is used to refer to a specific region of the user’s face defined prior to display of the enlarged stabilized image. The region of interest can be defined manually by the user or automatically by the device according to certain predefined criteria.

[0025] The term "image processing" refers to all techniques aimed at altering images in order to improve them. These may be adjusting contrast, sharpness and resolution, or applying filters and digital transformations, for example using machine learning models such as a convolutional neural network to optimize image quality and interpretation. Here, this definition also encompasses an alteration of the image perceived by the user that is not due to digital processing, in particular due to the presence of one or more optical components present on the device's screen, for example a refractive component such as one or more lenses. The term "compensating for a vision defect" should be understood to mean that the image processing at least partially corrects visual problems such as presbyopia, myopia or hyperopia by adapting the displayed image seen by the user to improve their visual perception. The user is thus able to better discern the details of the displayed image, which has been processed, than those of the same image without processing.

[0026] Device

[0027] The device can be a smartphone, tablet, laptop, or smart mirror. The term "smart mirror", also called a digital mirror, is used here to refer to a mirror that integrates digital display capabilities and communication with other devices or networks, usually using wireless technologies such as WiFi, Bluetooth or other communication protocols. In the context of the invention, a smart mirror can be used to display additional information, receive commands from the user or interact with other smart devices. In particular, the smart mirror can have a reflective surface, in particular a semi -reflective metal layer, behind which a screen is arranged. The information or images displayed on the screen can thus be substituted or superimposed in the image reflected by the semi -reflective surface.

[0028] The device can be portable, in particular configured to be held in one hand. Alternatively, the device may not be held by the user, for example it may be arranged on a fixed support, such as a table or shelf, or may be fixed to a wall.

[0029] Camera

[0030] The camera can have one or more CMOS or CDD sensors. The camera can be configured to acquire colour or monochrome images.

[0031] The camera can have electromechanical stabilization means configured to stabilize the camera’s optics or sensor.

[0032] The camera can be a front camera so that the user can be observed when they are facing the device's screen.

[0033] The device can have multiple cameras, in particular having different focal lengths and / or distinct viewing axes. It may thus be possible to efficiently capture the user's face at different viewing angles or with different optical magnifications. Screen

[0034] The screen can be of any type, for example an LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode) or AMOLED (Active Matrix Organic Light-Emitting Diode) screen.

[0035] The screen can define a rectangular surface. The resolution of the screen can be high definition (HD), i.e. 1280 pixels by 720 pixels, full HD, i.e. 1920 pixels by 1080 pixels, 4K, i.e. 3840 pixels by 2160 pixels, or higher.

[0036] The screen can be substantially flat over its entire surface. Alternatively, the screen can be curved over at least part of its surface, or even over its entire surface.

[0037] The largest dimension of the screen can be less than 100 cm, better still less than 50 cm, in particular between 10 and 40 cm. The term “largest dimension” is used here to refer to the greatest distance connecting two points located on the surface of the screen.

[0038] The screen can be touch-sensitive.

[0039] Interface, memory and processor

[0040] The device can have an interface.

[0041] The interface can be configured to allow the user to carry out commands. In particular when the screen is touch-sensitive, the screen can be part of the interface.

[0042] The device can have a memory, in particular configured to store images from the camera and / or displayed on the screen.

[0043] The device can have one or more processors configured to carry out all or some of the steps of the method according to the invention.

[0044] In particular, the memory can store acquired images of the user’s face, in particular images acquired during a treatment or therapy applied to the user’s face.

[0045] Reference image and region of interest

[0046] The acquired images can comprise a reference image. The region of interest may be visible in the reference image. The region of interest can correspond to a specific region of the user’s face, which may not encompass the entire face.

[0047] The device can comprise a pre-saved list of regions of interest.

[0048] In particular, the region of interest can be a region of an eye of the user, in particular the eyebrow, the upper eyelid, the lower eyelid, the eyelashes, the contour of the eye or the corner of the eye. The region of interest can encompass both eyes of the user.

[0049] Alternatively, the region of interest can be the nose, one of the cheeks, the mouth, the forehead or the chin of the user. The region of interest can be the entire face of the user. The definition of the entire face as a region of interest is particularly suitable for the application of a treatment or therapy to the entire face.

[0050] The interface can allow the user to manually define the region of interest, in particular by selecting from a drop-down menu displayed on the screen or by selecting a sub-region of an acquired image displayed on the screen.

[0051] The drop-down menu displayed on the screen can contain at least one, better still at least two, or even at least three, or more than four, different regions of interest. It should be understood here that the user can select, for example, one of the eyes or the nose as the region of interest. The region of interest can be selected from the drop-down menu using one or more buttons present on the interface. If the screen is touch-sensitive, the region of interest can be selected by simply touching the screen with a finger.

[0052] A sub-region of the reference image can be selected using one or more buttons present on the interface. If the screen is touch-sensitive, the region of interest can be selected by simply touching and / or dragging one or more fingers over the screen. For example, the region of interest is a region of the reference image that the user zooms to.

[0053] The device can be configured to determine the region of interest automatically using image analysis.

[0054] For example, when an applicator of a cosmetic product, for example a mascara brush, an application tip, a stick, a sponge or a puff, is detected in the vicinity of the user’s face, the region of interest is defined as the region treated by the applicator. The term “detected in the vicinity of’ should be understood to mean that the distance between the applicator and the user’s face, evaluated on the basis of an image acquired by the camera, is less than a predefined threshold.

[0055] Alternatively, when an applicator of a cosmetic product is detected in the vicinity of the face of the user, the region of interest can be determined from the type of applicator and / or the type of cosmetic treatment normally carried out using the applicator.

[0056] The term "applicator" is used here to refer to any tool for applying a cosmetic product, but also by extension one or more fingers of one hand. The applicator can be a pencil, a brush, a stick, a foam, and so on.

[0057] The region of interest can be readjusted according to the magnitude of the movements of the user. For example, the region of interest can be expanded if the movements of the applicator have a magnitude greater than a first predefined magnitude threshold, or the region of interest can be reduced if the movements of the applicator have a magnitude less than a second predefined magnitude threshold. Readjustment of the region of interest during use provides a reactive device, improving the user's comfort.

[0058] Alternatively, the region of interest can be locked once defined. In particular, the device can be configured to continuously display images of the region of interest after it has been defined.

[0059] The region of interest can be redefined on the basis of a command from the user.

[0060] Images

[0061] At least some of the acquired images can comprise the region of interest.

[0062] At least some of the acquired images can be saved in the device's memory. Optionally, acquired images that do not have the region of interest can be ignored by the device and not saved. Images can be processed without saving, on the fly.

[0063] Acquired images can be acquired by the device at a predefined acquisition frequency, in particular greater than 24 Hz, better still greater than 60 Hz, even better still greater than 90 Hz, or greater than 120 Hz.

[0064] The acquisition frequency is preferably high enough to obtain a smooth view for the user. Each of the displayed images corresponds to an acquired image and the acquired image can in particular be enlarged, and / or restored, and / or processed before being displayed.

[0065] Each of the displayed images can comprise the region of interest.

[0066] The display of images can vary according to the size of the device's screen. For example, with a large screen in the case of a smart mirror, it is possible to display the whole of the user's face while maintaining a sufficient level of detail in the region of interest.

[0067] The displayed images can be in colour or black and white.

[0068] Stabilization

[0069] At least one of the displayed images, in particular each of the displayed images, can be stabilized.

[0070] The stabilized image(s) can be saved in the device's memory.

[0071] The acquired images can comprise at least one reference image. The device can be configured to stabilize the displayed image by restoring the corresponding acquired image in relation to the reference image.

[0072] The term “restoring the acquired image in relation to the reference image” should be understood to mean that the acquired image is recentred by taking the reference image as the reference. Restoration keeps the centring of the images the same as that of the reference image. Acquired images restored in this way are hereinafter referred to as "restored acquired images".

[0073] The restoration can be a translation of the image in two non-parallel directions of the acquired image. In particular, the restoration can be a translation in two perpendicular directions of the acquired image, in particular in a first horizontal direction and a second vertical direction. Alternatively, other geometric transformations can be modelled to obtain global or local deformations.

[0074] Restoration can maximize the pixel-by-pixel correspondence between the acquired image and the region of interest of the reference image by considering each pixel and / or its surroundings. Alternatively, points of interest can also be used to calculate the optimum transformation. Alternatively, restoration can be performed on the basis of at least one predetermined landmark on the face. In this case, restoration is performed not using a reference image but on the basis of at least one landmark on the face that is visible in the acquired images.

[0075] The predetermined landmark can be a physical landmark on the user’s face. For example, it can be one or more points on the nose, the mouth, the eyes, the ears, the facial contour or a combination of these.

[0076] For a given acquired image, the device can be configured to detect the predetermined physical landmark in the acquired image and to restore said image so that said landmark remains at a predefined position in the displayed image.

[0077] Alternatively, stabilization can be partially achieved on the basis of stabilization means directly integrated in the device's camera.

[0078] Stabilization can be performed before the acquired image is enlarged. As a variant, stabilization can be performed after the acquired image has been enlarged.

[0079] Enlargement

[0080] At least one of the acquired images, in particular each of the acquired images, can be enlarged. In particular, at least one of the restored acquired images, in particular each of the restored acquired images, can be enlarged.

[0081] At least one of the acquired images, in particular each of the acquired images, can be enlarged by a predefined scale factor, for example between 1 and 5, better still between 1 and 4, even better still between 1 and 3, the limit 1 being excluded.

[0082] The scale factor can be constant for all corresponding acquired image / displayed image pairs.

[0083] Alternatively, the scale factor may not be constant for all corresponding acquired and displayed images. For example, a non-constant scale factor can compensate for a relative movement of the user moving away from or toward the camera.

[0084] The enlarged image(s) can comprise the entire region of interest. Alternatively, the enlarged image(s) may comprise only a sub-region of the region of interest.

[0085] Image processing The device can be configured to perform an image processing on the enlarged images.

[0086] The image processing can be carried out prior to display of the image so as to compensate for a vision defect chosen from presbyopia and hyperopia.

[0087] The image processing can increase the resolution of the displayed image.

[0088] Each of the displayed images can be processed so as to increase its resolution.

[0089] The resolution of the displayed image can be increased using "super-resolution" algorithms.

[0090] The term "super-resolution" is used here to refer to a set of algorithms and techniques used to improve or increase the resolution of a given image, in particular to oversample it. "Superresolution" algorithms can in particular comprise conventional interpolation or machine learning, or even deep learning, methods.

[0091] Thanks to the increased resolution, the displayed images are sharper and more visible to the user, especially once enlarged.

[0092] The image processing can increase the contrast and / or sharpness of the displayed image.

[0093] The image processing can involve applying a digital filter that accentuates contours of the displayed image, or that utilizes a model based on a machine learning method, this model preferably involving the use of a convolutional neural network.

[0094] The convolutional neural network may have been trained so as to overcorrect an image so that this overcorrection, once degraded by the visual defect to be corrected, is close to the original image when the image overcorrected in this way is seen by the user with the vision defect..

[0095] A filter that accentuates the contours of the displayed image is, for example, a high- pass filter, or a non-linear masking (unsharp masking).

[0096] The applied image processing can be selected by the user from multiple types of image processing using the interface, or can be generated automatically on the basis of a description by the user of their vision defect.

[0097] Various types of image processing can be preprogrammed in the device's memory, or in any other medium, for example a remote server that is accessible by the device using communication means. An image processing is, for example, selected using a drop-down menu displayed on the screen.

[0098] The device can be configured to display one or more examples of different image processings on the screen. The user can thus choose the type of processing they deem optimum to compensate for their vision defect, for example.

[0099] The description by the user of their vision defect comprises, for example, the entry of information relating to the user’s vision defect using the interface. The information can comprise, for one eye, in particular for each eye, the type of vision defect, the correction value, the degree of astigmatism and / or the orientation of astigmatism.

[0100] The interface can be configured to allow the user to calibrate their vision defect, the interface having calibration means for adjusting the image processing to be applied. The term “calibrating their vision defect” should be understood to mean that the user can perform commands in real time to directly adjust the type of image processing applied. For example, the interface can have gauges, sliders for adjusting the image processing. The user can thus choose the correction they deem optimum.

[0101] Alternatively, the image processing can be performed by means of one or more optical components, in particular one or more lenses, arranged on the device's screen.

[0102] The image processing according to the invention can be adaptable for multiple types of users having different vision defects.

[0103] The device may be designed to take at least one image through a corrective lens of a frame of the user, and, on the basis of this image, to determine the image processing to be applied in order to compensate for a vision defect of the user.

[0104] The term "hyperopia" is used here to refer to a refractive disorder that causes blurred vision from far and near. Presbyopia is an optical anomaly related to the aging of the lens of the eye. It is therefore an inevitable phenomenon affecting all individuals.

[0105] The term "presbyopia" is used here to refer to an inevitable age-related visual defect that leads to the loss of elastic deformation capability of the lens of the eye. The power of near focusing is therefore impacted, and near vision is thus impaired.

[0106] The displayed image can contain at least one information informing the user about the treatment or therapy to be performed. The information informs the user, for example, about the region of the face to which to apply a cosmetic product, and / or about the type of motions of the applicator to be performed, and / or the intensity of the movement to be performed.

[0107] The information can contain a highlighted pattern.

[0108] The pattern can be semi-transparent.

[0109] The pattern can have uniform opacity. The term "uniform opacity" should be understood to mean that the pattern has the same degree of opacity (or transparency) over its entire surface. As a variant, the pattern can have variable opacity, the opacity being increased in one or more regions of the pattern requiring the application of a greater quantity of cosmetic product, and / or a greater intensity of application. The user can more easily identify how to apply a given cosmetic product.

[0110] The information displayed can contain the pattern for make-up.

[0111] For example, the information displayed can contain the pattern for make-up for an eyebrow and / or an eyelid and / or the contour of the eye.

[0112] The device can be configured to analyse an image of the made-up face of the user and to generate information that facilitates subsequent reproduction of the make-up

[0113] The device can be configured to generate said information on the basis of an acquired image of make-up applied when the device was last used, this acquired image being in particular stored in the device's memory.

[0114] The device can be configured to generate the information on the basis of an acquired image of a first make-up applied to a region of a first eye. The information can then be transformed, in particular by axial symmetry in relation to an axis of symmetry of the face, so as to inform the user about the motions to be performed to make up the other eye. The device thus makes it easier to maintain symmetry when applying make-up to multiple regions of the face.

[0115] Viewing method

[0116] The subject of the invention is also, independently of or in combination with the foregoing, a method for viewing a part of the face of a user, carried out by a device having at least one screen and a camera for continuously acquiring images of the face of the user, in particular a device as defined previously, the method comprising: - detecting a predefined region of interest of the face of the user,

[0117] - displaying an enlarged stabilized image of the region of interest.

[0118] The device used to carry out the method can have all or some of the characteristics described above.

[0119] Thus, it is possible to obtain an enlarged and stabilized visualization of a region of interest.

[0120] Advantageously, the method comprises the user who is viewing the displayed image applying a therapy or a cosmetic product to the region of interest.

[0121] Preferably, an image processing an image processing is carried out prior to display of the image, the displayed image being processed so as to compensate for a vision defect of the user.

[0122] The method advantageously involves a step of calibrating the vision defect, comprising selecting an image processing from multiple types of image processing preprogrammed in the device, or automatically generating an image processing on the basis of a description by the user of their vision defect.

[0123] The image processing is, for example, selected using a drop-down menu displayed on the screen.

[0124] The calibration can involve displaying one or more examples of different image processings on the screen. The example(s) of image processings can comprise application of the image processing(s) to the same given acquired image.

[0125] The description by the user of their vision defect can comprise the entry of information relating to the user’s vision defect using the interface. The information can comprise, for one eye, in particular for each eye, the type of vision defect, the correction value, the degree of astigmatism and / or the orientation of astigmatism.

[0126] The method can involve continuously acquiring images, acquired by the camera, of the region of interest of the user’s face.

[0127] Cosmetic treatment method

[0128] The subject of the invention is also a cosmetic treatment method, comprising: - displaying an enlarged stabilized image of a region of interest by carrying out the aforementioned method,

[0129] - the user who is viewing the displayed image applying a therapy or a cosmetic product to the region of interest.

[0130] An image processing can be carried out prior to display of the image, the displayed image being processed so as to compensate for a vision defect of the user.

[0131] Computer program product

[0132] The subject of the invention is also, independently of or in combination with the foregoing, a computer program product intended to be used by a device having at least one screen, a processor, a memory and a camera for continuously acquiring images of the face of a user, this computer program product comprising code instructions that, when the program is executed by the processor, cause the processor to:

[0133] - detect a predetermined region of interest in a reference image acquired by the camera,

[0134] - display an enlarged stabilized image of the region of interest on the screen, and preferably

[0135] - carry out an image processing prior to display of the image so as to compensate for a vision defect of the user.

[0136] Brief description of the drawings

[0137] The invention will be better understood by reading the following detailed description of nonlimiting exemplary embodiments thereof and by examining the appended drawing, in which:

[0138] [Fig. 1] Figure 1 schematically and partially shows a user using an example device according to the invention,

[0139] [Fig. 2] Figure 2 schematically and partially shows the step of displaying a processed image to compensate for a vision defect of the user,

[0140] [Fig. 3] Figure 3 shows steps of an example method according to the invention,

[0141] [Fig. 4a] Figure 4a schematically and partially shows examples of patterns for making up the eyebrows, [Fig. 4b] Figure 4b schematically and partially shows examples of patterns for making up the eyelids, and

[0142] [Fig. 4c] Figure 4c schematically and partially shows examples of patterns for making up the contour of the eyes.

[0143] Detailed description

[0144] Figure 1 shows a user U using the device 1 according to the invention. As shown, the device 1 can be a smartphone, but it could be any other suitable device.

[0145] The screen 2 can be rectangular, as shown.

[0146] This screen 2 can be touch-sensitive.

[0147] In the example under consideration, the device 1 has a front camera 3.

[0148] In the example shown, the user U applies a treatment or therapy to the region of the eye using an applicator 4.

[0149] The device 1 is configured to acquire a sequence of images of the face of the user U.

[0150] The reference image Iref is, for example, an image of the entire face of the user.

[0151] In this example, the region of interest Iroiis the region of the eye being treated.

[0152] The region of interest Irefcan be selected manually by the user U in the reference image Iref displayed on the screen 2.

[0153] The device 1 is configured to enlarge the reference image Iref in the region of interest Iroi-

[0154] Stabilization is performed so as to maintain a relatively static appearance for the region of interest Iroiin the images displayed on the screen 2.

[0155] Figure 2 shows an example of processing provided for the displayed images.

[0156] The top left image Al is an enlarged representation of the region of interest Iroias displayed on the device 1 without corrective processing, while the bottom left image A2 is an enlarged representation of the region of interest Iroias perceived by the user U with their vision defect.

[0157] The top right image A3 is an enlarged representation of the corrected region of interest Iroias displayed on the device 1 with correction. The features of the image can be accentuated by increasing the contrast and the resolution, in particular by using a super-resolution algorithm. The bottom right image is an enlarged representation of the corrected region of interest Iroias perceived by the user U with their vision defect.

[0158] It is clear from Figure 2, by comparing images A2 and A4, that the perception of details by the user U is improved by virtue of the invention.

[0159] Figure 3 is a schematic representation of steps of an example method according to the invention.

[0160] In step 10, images are continuously acquired by the camera 3 of the device 1.

[0161] The method can comprise a possible step 12 of calibrating the vision defect of the user U. To do this, the user U selects, for example, the image processing that best compensates for their vision defect from multiple examples of image processings applied to one or more images displayed on the screen 2. The images displayed on the screen can be images acquired by the camera 3 in step 10.

[0162] In this example, the method comprises a step 14 of defining a reference image Iref and a region of interest Iroi. For example, the user U manually selects the region of interest Iroiby zooming in on an image displayed on the screen 2, the displayed image then being taken as the reference image Iref.

[0163] In step 16, each acquired image is restored in relation to the reference image Iref, for example by maximizing the alignment of the acquired image with the region of interest Iroiof the reference image Iref. The restored acquired image can be saved in the device's memory. Restoration of the acquired image thus allows the image displayed on the screen 2 in step 20 to be stabilized.

[0164] In step 18, the restored acquired image is enlarged, for example by digital zooming.

[0165] The image enlarged in step 18 can be processed so as to compensate for a vision defect of the user U, chosen from presbyopia, hyperopia and myopia, before it is displayed. The applied image processing can, alternatively or simultaneously, increase the contrast and / or sharpness of the displayed image.

[0166] It is possible to embed information that helps the user U to carry out a cosmetic treatment, for example making up, in the displayed processed image. This information is, for example, in the form of a highlighted pattern 30,32,34 displayed in the image. Figure 4a shows different examples of make-up patterns for the eyebrows 30 that can be displayed.

[0167] Figure 4b shows different make-up patterns for the upper eyelids 32.

[0168] Figure 4c shows different make-up patterns for the contours of the eyes 34.

[0169] Of course, the invention is not limited to the examples that have just been described. The invention can be used for correcting vision defects other than those stated.

Claims

Claims1. Device (1) for viewing a part of the face of a user (U), having at least one screen (2) and a camera (3) for continuously acquiring images of the face of the user (U) so as to allow the user to view images displayed by the screen when they are facing the camera (2), the acquired images comprising a region of interest (Iroi) of the face of the user (U), the device (1) being configured to generate, on the basis of the acquired images, enlarged stabilized images of the region of interest (Iroi) and to display them on the screen (2), an image processing being carried out prior to display of the images, the images displayed on the screen (2) being processed so as to at least partially compensate for at least one vision defect of the user (U), in particular, a defect chosen from presbyopia and hyperopia.

2. Device (1) according to the preceding claim, the image processing increasing the resolution and / or contrast and / or sharpness of the displayed image.

3. Device (1) according to any one of the preceding claims, the image processing involving applying a digital filter that accentuates contours of the displayed image, or involving a model based on a machine learning method aimed at compensating for the vision defect of the user (U), this model preferably involving the use of a convolutional neural network.

4. Device (1) according to any one of the preceding claims, the device (1) having an interface (3), the applied image processing being selected by the user (U) from multiple types of preprogrammed image processing using the interface (3), or being generated automatically on the basis of a description by the user (U) of their vision defect.

5. Device (1) according to Claim 4, the interface (3) being configured to allow the user (U) to calibrate their vision defect, the interface (3) having calibration means for adjusting the image processing to be applied.

6. Device (1) according to any one of the preceding claims, the device (1) having an interface (3) allowing the user (U) to manually define the region of interest (Iroi), in particular by selecting from a drop-down menu displayed on the screen (2) or by selecting a sub-region of an acquired image displayed on the screen (2).

7. Device (1) according to any one of the preceding claims, the device (1) being configured to determine the region of interest (Iroi) automatically using image analysis.

8. Device (1) according to the preceding claim, wherein when an applicator (4) of a cosmetic product is detected in the vicinity of the face of the user (U), the region of interest (Iroi) is defined as the region treated by the applicator (4).

9. Device (1) according to Claim 7, wherein when an applicator (4) of a cosmetic product is detected in the vicinity of the face of the user (U), the region of interest (Iroi) is determined from the type of applicator (4) and / or the type of cosmetic treatment normally carried out using the applicator (4).

10. Device (1) according to any one of the preceding claims, the acquired images comprising at least one reference image (Iref), the device being configured to stabilize the displayed image by restoring the corresponding acquired image in relation to the reference image (Iref)-11. Device (1) according to the preceding claim, the restoration maximizing the pixel-by-pixel correspondence between the acquired image and the region of interest (Iroi) of the reference image (Iref), or the restoration being performed on the basis of at least one predetermined landmark on the face.

12. Device (1) according to any one of the preceding claims, the displayed image containing at least one information informing the user about a treatment or therapy to be performed, this information preferably containing a highlighted pattern (32, 33, 34).

13. Device (1) according to any one of the preceding claims, the device (1) being configured to analyse an image of the made-up face of the user (U) and to generate information that facilitates subsequent reproduction of the make-up.

14. Device (1) according to any one of the preceding claims, the region of interest (Iroi) being a region of an eye of the user (U).

15. Device according to any one of the preceding claims, the vision defect being chosen from presbyopia and hyperopia.

16. Method for viewing a part of the face of a user (U), carried out by a device (1) having at least one screen (2) and a camera (3) for continuously acquiring images of the face of the user (U), in particular a device (1) according to any one of the preceding claims, the method comprising:- detecting a predefined region of interest (Iroi) of the face of the user (U),- displaying an enlarged stabilized image of the region of interest (Iroi).

17. Cosmetic treatment method, comprising:- displaying an enlarged stabilized image of a region of interest (iroi) by carrying out the method according to Claim 16,- the user (U) who is viewing the displayed image applying a therapy or a cosmetic product to the region of interest (Iroi).

18. Method according to either of the two preceding claims, an image processing being carried out prior to display of the image, the displayed image being processed so as to compensate for a vision defect of the user (U).

19. Computer program product intended to be used by a device (1) having at least one screen (2), a processor, a memory and a camera (3) for continuously acquiring images of the face of a user (U), in particular a device (1) according to any one of Claims 1 to 15, the computer program product comprising code instructions that, when the program is executed by the processor, cause the processor to:- detect a predetermined region of interest (Iroi) in a reference image (Iref) acquired by the camera (3),- display an enlarged stabilized image of the region of interest (Iroi) on the screen (2), and preferably- carry out an image processing prior to display of the image so as to compensate for a vision defect of the user (U).

Images