METHOD AND DEVICE FOR DISPLAYING DATA FOR MONITORING AN EVENT

Description

TECHNICAL FIELD OF THE INVENTION

[0001] The field of the invention is that of digital data processing.

[0002] More specifically, the invention relates to a method and device for displaying data for monitoring an event.

[0003] The invention finds applications in particular for the live monitoring of a sporting event, such as a football, rugby, basketball, tennis, etc. match, in a stand of a sports facility such as a stadium or a hall. The invention also finds applications in the field of entertainment, for example for monitoring a game, a live show or a concert. STATE OF THE ART

[0004] Techniques are known from the prior art for monitoring an event by displaying, in particular in real time, data, such as statistics relating to an individual or a group of individuals participating in the event, such as, for example, the number of goals scored by a player during a football match, the number of aces or unforced errors made by a player during a tennis match, the success rate of a player's three-point shots during a basketball match, etc.

[0005] Such data is usually displayed on a screen of the equipment in which the event takes place.

[0006] The major drawback of these techniques is that they are not very intuitive for an individual following the event from a stand on the equipment. In addition, these techniques tend to distract the attention of the individual who must turn his head to look at a screen on the equipment.

[0007] None of the current systems can simultaneously meet all the required needs, namely to offer a technique that allows an individual to follow an event by intuitively displaying data associated with the actors of the event without distracting the individual's attention.

[0008] US Patent 2010 / 026801 (WILLIAMS MICHAEL JOHN [GB] ET AL) discloses a method and apparatus for generating a log of events occurring during a sporting event. In particular, the method discloses the playback of player information overlays in a video captured from a sports field using a stationary camera.

[0009] The article entitled "A Vision-Based Mobile Augmented Reality System for Baseball Games", SEONG-OH LEE ET AL, (2011-07-09), INTERNATIONAL CONFERENCE ON COMPUTER ANALYSIS OF IMAGES AND PATTERNS, PAGE(S) 61 - 68, presents a portable augmented reality system for baseball games, and discloses a method performing element overlays in an acquired image of a baseball field, the overlay representing information associated with a player, where the information is obtained by the system via wireless communication with an external information management system. STATEMENT OF THE INVENTION

[0010] The present invention aims to remedy all or part of the drawbacks of the state of the art cited above.

[0011] To this end, the invention relates to a real-time augmented reality method, comprising steps of: acquisition of a plurality of images by an image acquisition device at least partially covering a space, the space having at least two landmarks, the image acquisition device being associated with a two-dimensional reference frame, called the image reference frame, the image acquisition device being included in a portable electronic device also comprising a screen; detection of at least two landmarks of the space in at least one image, the space being associated with a three-dimensional reference frame, called the space reference frame; determination of a three-dimensional position and orientation of the space relative to the image acquisition device using the detected landmarks; reception of the instantaneous position, in the space reference frame, of a mobile element moving in the space;calculating the position of the mobile element in the image frame from the transformation parameters between the space frame and the image frame, said transformation parameters being calculated from the position and the three-dimensional orientation of the space relative to the image acquisition device; displaying in real time at least one acquired image on the screen; superimposing in real time on the image displayed on the screen at least one inlay; a predetermined distance from the position of the moving element in the image frame.

[0012] Thus, by knowing the transformation between the spatial reference point and the image reference point, it is possible to display data associated with said mobile element participating in a live event such as a sports match, a show or a concert, in the overlay displayed near the image of the mobile element on the screen retransmitting the image acquired by the image acquisition device.

[0013] The space may include, for example, a match field or a stage of a performance hall. The space is generally delimited to allow spectators to follow the live event taking place in the space, for example from at least one stand near the field or the stage.

[0014] The moving element is usually an actor participating in the live event, such as a player in a field sport match, an actor in a show, or a musician in a concert. The moving element can also be a prop, such as a game prop played by players during a match. A game prop is usually a ball, a puck, or a shuttlecock.

[0015] It should be noted that the overlay allows you to display data, a static or animated image, a video, or any other element that enhances the image of the event displayed on the screen.

[0016] The transformation parameters between the space reference frame which is three-dimensional and the image reference frame which is two-dimensional are generally calculated from the three-dimensional position of the space relative to the image acquisition device and from the three-dimensional orientation of the space relative to the image acquisition device which may be a camera.

[0017] It should be noted that the image acquisition device can be represented in the spatial frame. This representation generally includes the three-dimensional coordinates of the image acquisition device in the spatial frame and the three angles used to orient the image acquisition device in the spatial frame.

[0018] The transformation between the spatial reference frame and the image reference frame generally includes at least one translation, at least one rotation and one projection.

[0019] It should be emphasized that the determination of the three-dimensional position and orientation of the space relative to the image acquisition device, in a frame associated with the image acquisition device or directly in the frame of the space, is carried out by detecting landmarks in an image covering at least partially the space and not by using a depth-of-field camera. Indeed, given the generally very large distance between the image acquisition device and the space, a depth-of-field camera would not be suitable because it induces too many inaccuracies in the position and orientation of the space.

[0020] Advantageously, a bitter is chosen from: a line of a marking of a terrain or a plateau included in the space; an arc of a circle of a marking of a terrain or a plateau included in the space; an intersection between two lines of a marking of a land or plateau included in the space; an element standing substantially perpendicular to the surface of a land or plateau included in the space; a characteristic element of a structure surrounding the surface of a land or plateau included in the space; a logo; a marker.

[0021] Preferably, four landmarks are detected and used to determine the three-dimensional position and orientation of the space relative to the image acquisition device.

[0022] In particular embodiments of the invention, the augmented reality method also comprises a step of automatic recognition of the type of terrain included in the space.

[0023] This automatic field type recognition step is generally based on the detection of characteristic points related to the shape of the recognized sports field which can be of any type: football, basketball, handball, rugby, tennis, hockey, baseball, etc. These characteristic points can be confused with the detected landmarks. It should be emphasized that this recognition step is not specific to a particular sport but allows to recognize any sports field whose typical characteristic points are known. The characteristic points are generally the general shape of the field, the relative position of the lines in relation to the field, the presence and relative position of arcs in relation to the field, etc.

[0024] Advantageously, the automatic recognition of the type of terrain is carried out by means of a deep learning method trained on a plurality of terrain images.

[0025] This makes it possible to quickly recognize any type of terrain, regardless of its orientation or viewing angle.

[0026] Furthermore, thanks to this deep learning method, also called by the English term " Deep Learning”, It is possible to recognize a terrain from a partial image of the terrain, that is, without needing to see the terrain as a whole.

[0027] In particular embodiments of the invention, the augmented reality method also comprises steps of: acquiring an instantaneous movement of the image acquisition device, in rotation and translation relative to space; updating the position and orientation of the sports field relative to the image acquisition device from the previous position and orientation of space relative to the image acquisition device and the instantaneous movement of the image acquisition device.

[0028] Thus, the data overlays in the images displayed on the screen are more stable. In addition, these additional steps make it possible to obtain an augmented reality process that consumes less computing time and therefore less electrical energy. Indeed, once the three-dimensional position and orientation of the space relative to the image acquisition device are known, it is easy to update them by knowing the movements of the image acquisition device. One method that can be used to evaluate these movements is, for example, SLAM (an acronym for the English term "SLAM"). Simultaneous Localization And Mapping ").

[0029] In particular embodiments of the invention, the step of determining the three-dimensional position and orientation of the space relative to the image acquisition device comprises a sub-step of generating parameters of a machine learning computer algorithm from a plurality of images recorded in a database, each image of the database representing all or part of a space whose position and orientation relative to the image acquisition device having acquired said image are known.

[0030] Thus, the determination of the three-dimensional position and orientation of the terrain relative to the image acquisition device can be carried out quickly and accurately by using the parameters generated in the acquired images to determine the three-dimensional position and orientation of the space.

[0031] In particular embodiments of the invention, the step of determining a three-dimensional position and orientation of the space relative to the image acquisition device comprises a sub-step of superimposing a three-dimensional model of the space on at least one of the images acquired by the image acquisition device.

[0032] Thus, the space reference can be positioned and oriented in virtual space relative to the image acquisition device.

[0033] In particular embodiments of the invention, the augmented reality method also comprises a step of correcting the instantaneous position of the mobile element as a function of an instantaneous speed of the mobile element and / or an instantaneous acceleration of the mobile element.

[0034] Thus, it is possible to improve the position of the inlay in the images, particularly when there is a significant latency between the acquisition of the image and its display. Indeed, the use of the instantaneous speed of the mobile element and / or the instantaneous acceleration of the latter makes it possible to predict a position of the mobile element in a close time interval.

[0035] The superimposition of the overlay on the image displayed on the screen of at least one piece of data associated with said mobile element is carried out in real time.

[0036] It should be noted that the data embedded in the image is usually transmitted by a data provider.

[0037] Advantageously, the inlay includes at least one piece of data chosen from: a name of a player; a statistic associated with the player, such as a number of goals, a number of tries, a number of baskets, a number of points scored, a number of successful passes; a name of a team; a positioning of a group of players in relation to other players; a formation of the team or a group of players; a distance between a point on the field and a player; a distance between two points on the field; and / or a graphic element such as a line, a circle, an ellipse, a curve, a square or a triangle; and / or a still or animated image; and / or a video.

[0038] An animation can, for example, be linked to the celebration of a score during a match.

[0039] In particular embodiments of the invention, the augmented reality method comprises a step of determining a clipping of a mobile element, the clipping generating an occlusion for at least one inlay superimposed on the image displayed on the screen.

[0040] Thus, it is possible to have a much more realistic rendering by hiding part of the overlay displayed at the level of a moving element, and in particular at the level of a player. This is particularly the case when the overlay includes a graphic element such as a virtual line on the field, corresponding for example to an offside line in football or rugby.

[0041] In particular embodiments of the invention, the augmented reality method also comprises a step of selecting a mobile element and displaying information relating to the mobile element in an overlay near the mobile element.

[0042] The invention also relates to a portable electronic device comprising a camera and a screen, implementing the augmented reality method according to any one of the preceding embodiments.

[0043] The portable electronic device also generally includes a processor and computer memory storing instructions for a computer program implementing the augmented reality method.

[0044] Preferably, the portable electronic device is a smart mobile phone, also known by the English term "smartphone", an augmented reality goggle or an augmented reality headset.

[0045] The wearable electronic device may include a frame and a display mounted on the frame that is intended to be worn on an individual's face.

[0046] In other words, the portable electronic device may include any means of reproducing an image that can be displayed in front of an individual's eye, including a contact lens for reproducing an image.

[0047] In particular embodiments of the invention, the portable electronic device also comprises at least one accelerometer and / or a gyroscope.

[0048] Thus, the device includes means for evaluating the translational and rotational movements of the camera relative to space.

[0049] It should be noted that part of the process could be implemented by a remote server, including the steps of: detection of at least two landmarks of the terrain in at least one image, the space being associated with a three-dimensional reference frame, called the space reference frame; and determination of a three-dimensional position and orientation of the space relative to the image acquisition device using the detected landmarks.

[0050] In which case, the step of updating the three-dimensional position and orientation of the space based on the evaluation of the camera's movements relative to the space is implemented by the portable electronic device. BRIEF DESCRIPTION OF THE FIGURES

[0051] Other advantages, aims and particular characteristics of the present invention will emerge from the following non-limiting description of at least one particular embodiment of the devices and methods which are the subject of the present invention, with reference to the appended drawings, in which: [ Fig 1 ] there figure 1is a block diagram of an augmented reality method according to the invention; [ Fig 2 ] there figure 2 is a view of a portable electronic device implementing the augmented reality method of the figure 1 . DETAILED DESCRIPTION OF EMBODIMENTS OF THE INVENTION

[0052] This description is given without limitation, each characteristic of an embodiment being able to be combined with any other characteristic of any other embodiment in an advantageous manner.

[0053] Please note, from now on, that the figures are not to scale. Example of a particular embodiment of the invention

[0054] There figure 1 illustrates a block diagram of an augmented reality method 100 according to the invention implemented by a portable electronic device 200 illustrated in figure 2 .

[0055] The portable electronic device 200 is here a smart mobile phone held by an individual (not shown) installed in a stand of a space 202 which is a room where a basketball game is taking place between two teams each comprising five players 230. The two teams compete for a game accessory which is a basketball (not shown). The game takes place on a basketball court 220, comprising a marking 221. The room 202 corresponds in this case to a space, the space comprising the court 220 and structures such as the stand and two basketball baskets 203 (only one shown in the figure 2 ).

[0056] In this example, the individual uses the mobile phone 200 and sees on a screen 250 of the mobile phone 200 the image acquired in real time acquired by an image acquisition device 210 which is a camera. The camera 210 here captures a part of the space 202, including in particular a part of the basketball court 220. In the situation illustrated in figure 2 , the players 230 of the two teams are located in one half of the court, one of the two teams, represented by the horizontal stripes is attacking while the other team, represented by the vertical stripes is defending, that is to say preventing the players 230 1 of the team in attacking position from sending the basketball into the basketball basket 203.

[0057] The method 100 thus comprises a first step 110 of acquiring a plurality of images by the camera. It should be emphasized that the acquired images generally form a video stream.

[0058] In the field of view of the camera 210, four landmarks 222 are detected during the second step 120 of the method. The four landmarks 222 are a corner 222 1 of the court, the basketball basket 203, an arc 222 3 representing the three-point line and an arc 222 4 surrounding a free throw line. It should be noted that the corner 222 1 , the arc 222 3 and the arc 222 4 are part of the marking 221 of the basketball court.

[0059] Optionally, the method 100 may comprise a step 115 prior to or concomitant with the step 120 of detecting the landmarks 222, during which the type of terrain is recognized by means of a terrain recognition algorithm based on deep learning (“ Deep learning") trained on a plurality of sports field images. The algorithm can in particular recognize whether it is a basketball field, a football field, a rugby field, a hockey field, a tennis court or any other field comprising a plurality of landmarks. It should be noted that the landmarks detected during step 120 are generally a function of the type of field detected.

[0060] Thanks to the landmarks 222 detected in the space 202, a three-dimensional position and orientation of the space 202 relative to the camera 210 are determined during a third step 130 of the augmented reality method.

[0061] The determination of the three-dimensional position and orientation can be carried out either by superimposing on the image a model of the space 202 using the landmarks 222, or by using parameters of a computer algorithm of machine learning also called by the English term " machine learning.

[0062] The model of the space 202 generally comprises a model of the terrain 220 and the marking 221, or even a model of singular elements which can act as landmarks such as, for example, basketball baskets 203. To superimpose the landmarks 222 detected in the image and the landmarks present in the model, a homography method or a method of the type " Perspective-n-Point » can be used.

[0063] When a machine learning algorithm is used, the method generally comprises an additional step of generating parameters of the algorithm from images of space 202 previously recorded in a database, each image being recorded with the three-dimensional position and orientation of an image acquisition device having acquired said image. This learning step, requiring a lot of computing time, is generally carried out by a remote server. It should be emphasized that this learning step is generally carried out only once.

[0064] Furthermore, the position of the space 202 relative to the camera 210 is generally calculated in a reference frame which is associated either with the camera 210 or with the space 202. The transition from the camera reference frame to the space reference frame is generally easily carried out by a translation and a rotation, these two reference frames being three-dimensional.

[0065] Transformation parameters between the spatial reference frame and a two-dimensional reference frame associated with the camera 210, called the image reference frame, are calculated to transform the coordinates obtained in the spatial reference frame into the reference frame of the images obtained by the camera 210. It should be emphasized that the image reference frame is distinct from the camera reference frame in the sense that the image reference frame is a two-dimensional reference frame and the camera reference frame is a three-dimensional reference frame. Generally, a projection makes it possible to move from the camera reference frame to the image reference frame.

[0066] The instantaneous position of a mobile element 235 in the spatial reference frame is then received during the fourth step 140 of the augmented reality method 100. The mobile element 235 is here one of the five players 230 of one of the two teams competing during the basketball game. The mobile element can also be the game accessory contested by the two teams, namely the basketball (not shown in the figure 2 ).

[0067] A calculation of the position of the mobile element 235 in the image frame of reference is carried out during the fifth step 150 of the augmented reality method 100.

[0068] Since the position of the mobile element 235 is known in the image frame, it is possible to superimpose on the image displayed on the screen 250 an inlay 240 near the position of the mobile element during a sixth step 160 of the method 100. Generally, the position of the inlay 240 is carried out at a predetermined distance from the instantaneous position of the mobile element 235 in the image. In the present non-limiting example of the invention, the inlay 240 is superimposed vertically relative to the position of the mobile element 235 so that it appears above the mobile element 235 on the displayed image.

[0069] It should be emphasized that it is possible, as on the figure 2 , to repeat steps 140 to 160 in order to display inlays 240 for a plurality of mobile elements present in space 202, here for the five players 230 1 of the team which is in the attack phase.

[0070] The overlay 240 for each player 230 1 may include one or more data, such as the player's name and the number of points scored since the start of the match. Any other statistical data useful for monitoring the match may be displayed with this method 100. The overlay 240 may also include an animation displayed as soon as the mobile element 235 has scored, by sending the ball into the basketball basket of the opposing team.

[0071] Thus, the individual in a stand can watch the basketball game while consulting the data of the players 230 1 on the screen of his telephone 200. It should be emphasized that the image displayed on the screen 250 of his telephone 200 can advantageously be superimposed on the field seen by the individual, so that the individual can follow the game without loss of attention. In addition, the individual is not obliged to turn his head to look at a screen (not shown in the figure 2) present in room 202.

[0072] In addition, because the data is displayed directly in the vicinity of the players 230, tracking is more intuitive. Using the screen 250, which is usually touch-sensitive, the individual can also select the type of data they wish to view, such as a particular player statistic.

[0073] It should be emphasized that the data overlay during the method 100 is performed in real time with respect to the acquired image. In other words, the calculation time between the acquisition of the image and the display thereof with the overlay(s) 240 is performed in a very short period of time, generally less than one millisecond, such that the individual can see the acquired images of the event almost simultaneously with a direct view of the event.

[0074] For this purpose, steps 120 and 130, which are computationally intensive, can advantageously be carried out on a remote server (not shown in the figure 2 ). At least one acquired image is thus transmitted to the remote server via telecommunication means (not shown in the figure 2 ) included in the 200 phone.

[0075] In order to reduce latency, the transmission of data between the telephone 200 and the remote server may be carried out using a telecommunications network configured according to the 5G standard. In addition, latency may also be reduced by using a computer server close to an antenna of the telecommunications network, the computer server then acting as the remote server performing the calculations of steps 120 and 130. In this case, this type of architecture is known as " Edge Computing”.

[0076] From the three-dimensional position and orientation of the terrain relative to the camera 210 calculated by the remote server for a given image, the method 100 updates this position and this orientation according to the movements of the camera 210 in the three-dimensional space during a step 170 replacing the steps 120 and 130, using for example a SLAM method (acronym for the English term "SLAM"). Simultaneous Localization And Mapping ").

[0077] These movements are for example acquired by a three-axis accelerometer 260 included in the telephone 200 during a step 175 carried out prior to step 170.

[0078] Thus, the calculation time being faster, the display of the data associated with the players is more stable compared to the instantaneous position of the mobile elements 230.

[0079] Also with the aim of improving the position of the inlay 240 of the data, in particular in order to give the impression that the mobile element 235 is followed by the method 100, the method 100 may comprise an optional step 180 included before the step 160 in order to correct the instantaneous position of the mobile element 230 in the spatial reference frame as a function of an instantaneous speed and / or an instantaneous acceleration of the mobile element 230. This instantaneous speed and this instantaneous acceleration, provided for example with the data associated with the mobile elements or calculated from successive instantaneous positions, make it possible to predict the position of the mobile element 230 in a short time interval.

[0080] This step 180 makes it possible in particular to overcome latencies which may occur in the transmission of data to the telephone 200.

[0081] In order to improve the realism of the display, in particular when at least one graphic element such as a line, a circle, a triangle is displayed, the method may comprise a step 190 during which an occlusion is calculated from a clipping of a mobile element, such as a player. This occlusion makes it possible to remove a part of the graphic element embedded on the screen or to prevent the embedding of this part of the graphic element coming to be superimposed on the mobile element. The occlusion step 190 may be carried out before or after the embedding step 160.

[0082] Knowing the position of a mobile element at a given instant in the image, the clipping can be carried out by means of a detection of an outline of this element or by any other techniques known to those skilled in the art. Another clipping technique consists of an estimation of the pose of a model representative of a mobile element, such as a player. Thus, knowing the usual size of a player, an estimation of the overall posture of the player can be carried out by analyzing the visible part of this player on an image or in a sequence of images, by detecting in particular characteristic points of the structure of the player, generally points of articulation of the skeleton of the player. By estimating the overall posture of the player, it is then possible to define an estimation of the total volume occupied by the player as well as his position in space.

[0083] From the position of the terrain and the moving elements in the image, it is possible to select, during an optional step 195, a particular moving element by clicking or touching an area of ​​the screen, called the interaction area, close to the image of the moving element. From the coordinates of the interaction area with the screen, it is possible to display in an overlay the statistics of the player closest to the coordinates of the interaction area.

Claims

1. Augmented reality method (100) in real time, comprising steps of: - acquiring (110) a plurality of images by an image acquisition device (210) that at least partially cover a space (202), the space having at least two landmarks (222), the image acquisition device being associated with a two-dimensional reference frame, referred to as the reference frame of the image, the image acquisition device being comprised in a portable electronic device (200) also comprising a screen (250); - detecting (120) at least two landmarks (222) of the space in at least one image, the space being associated with a three-dimensional reference frame, referred to as the reference frame of the space; - determining (130) a three-dimensional position and orientation of the space in relation to the image acquisition device thanks to the landmarks detected; - receiving (140) the instantaneous position, within the reference frame of the space, of a mobile element (230) moving in the space; - calculating (150) the position of the mobile element (235) in the reference frame of the image from transformation parameters between the reference frame of the space and the reference space of the image, said transformation parameters being calculated from the three-dimensional position and orientation of the space in relation to the image acquisition device; - displaying in real time at least one acquired image on the screen (250); - superimposing (160) on the image displayed on the screen in real time at least one overlay (240) at a predetermined distance in relation to the position of the mobile element (235) in the reference frame of the image.

2. Augmented reality method according to claim 1, wherein a landmark (222) is chosen from: - a line of a marking of a field or of a stage comprised in the space; - a semi-circle of a marking of a field or of a stage comprised in the space; - an intersection between two lines of a marking of a field or of a stage comprised in the space; - an element standing substantially perpendicularly in relation to the surface of a field or of a stage comprised in the space; - an element characteristic of a structure surrounding the surface of a field or of a stage comprised in the space; - a logo; - a marker.

3. Augmented reality method according to any of claims 1 to 2, also comprising a step (115) of automatic recognition of the type of a field comprised in the space.

4. Augmented reality method according to claim 3, wherein the automatic recognition of the type of a field is carried out via a method of deep learning trained on a plurality of field images.

5. Augmented reality method according to any of claims 1 to 4, also comprising steps of: - acquiring (175) an instantaneous movement of the image acquisition device, in rotation and in translation in relation to the space; - updating (170) the position and the orientation of the space in relation to the image acquisition device from the preceding position and orientation of the space in relation to the image acquisition device and of the instantaneous movement of the image acquisition device.

6. Augmented reality method according to any of claims 1 to 5, wherein the step of determining the three-dimensional position and orientation of the space in relation to the image acquisition device comprises a substep of generating parameters of a computer machine learning algorithm from a plurality of images recorded in a database, each image of the database representing all or a portion of a space of which the position and the orientation in relation to the image acquisition device having acquired said image are known.

7. Augmented reality method according to any of claims 1 to 6, wherein the step of determining a three-dimensional position and orientation of the space in relation to the image acquisition device comprises a substep of superimposing a model of the space on at least one of the images acquired by the image acquisition device.

8. Augmented reality method according to any of claims 1 to 7, also comprising a step (180) of correcting the instantaneous position of the mobile element according to an instantaneous speed of the mobile element and / or of an instantaneous acceleration of the mobile element.

9. Augmented reality method according to any of claims 1 to 8, wherein the overlay (240) comprises at least one piece of data chosen from: - a name of a player; - a statistic associated with the player, such as a number of goals, a number of tries, a number of baskets, a number of points scored, a number of successful passes; - a name of a team; - a positioning of a group of players in relation to other players; - a formation of the team or of a group of players; - a distance between a point of the field and a player; - a difference between two points of the field ; and / or a graphic element such as a line, a circle, a square or a triangle; and / or a fixed or animated image; and / or a video.

10. Augmented reality method according to any of claims 1 to 9, comprising a step of determining a clipping of a mobile element, the clipping generating an occlusion for at least one overlay superimposed on the image displayed on the screen.

11. Augmented reality method according to any of claims 1 to 10, also comprising a step of selecting a mobile element and of displaying a piece of information relating to the mobile element in an overlay in the vicinity of the mobile element.

12. Portable electronic device (200) comprising a camera (210) and a screen (250), implementing the augmented reality method according to any of claims 1 to 11.

13. Portable electronic device according to claim 12, characterized in that it is a smartphone, augmented reality glasses or an augmented reality headset.

14. Portable electronic device according to any of claims 12 to 13, also comprising at least one accelerometer (260) and / or a gyroscope.