WO2018002533A1 - Procédé d'occultation d'un objet dans une image ou une vidéo et procédé de réalité augmentée associé - Google Patents
Procédé d'occultation d'un objet dans une image ou une vidéo et procédé de réalité augmentée associé Download PDFInfo
- Publication number
- WO2018002533A1 WO2018002533A1 PCT/FR2017/051744 FR2017051744W WO2018002533A1 WO 2018002533 A1 WO2018002533 A1 WO 2018002533A1 FR 2017051744 W FR2017051744 W FR 2017051744W WO 2018002533 A1 WO2018002533 A1 WO 2018002533A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image
- face
- generating
- model
- glasses
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 197
- 230000003190 augmentative effect Effects 0.000 title claims abstract description 33
- 239000011521 glass Substances 0.000 claims description 244
- 210000001508 eye Anatomy 0.000 claims description 55
- 230000001815 facial effect Effects 0.000 claims description 24
- 238000012545 processing Methods 0.000 claims description 15
- 238000011161 development Methods 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 14
- 239000003086 colorant Substances 0.000 claims description 13
- 230000009466 transformation Effects 0.000 claims description 13
- 238000011049 filling Methods 0.000 claims description 10
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000012986 modification Methods 0.000 claims description 9
- 230000004048 modification Effects 0.000 claims description 9
- 230000003287 optical effect Effects 0.000 claims description 8
- 238000002360 preparation method Methods 0.000 claims description 2
- 210000000744 eyelid Anatomy 0.000 description 14
- 210000003128 head Anatomy 0.000 description 13
- 230000006870 function Effects 0.000 description 12
- 230000018109 developmental process Effects 0.000 description 10
- 210000004209 hair Anatomy 0.000 description 10
- 239000011159 matrix material Substances 0.000 description 8
- 238000011282 treatment Methods 0.000 description 8
- 230000014509 gene expression Effects 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 6
- 238000001514 detection method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000002093 peripheral effect Effects 0.000 description 5
- 238000005070 sampling Methods 0.000 description 5
- 239000013598 vector Substances 0.000 description 5
- 230000000007 visual effect Effects 0.000 description 5
- 238000013459 approach Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 239000000463 material Substances 0.000 description 4
- 238000013519 translation Methods 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 3
- 238000004737 colorimetric analysis Methods 0.000 description 3
- 210000000887 face Anatomy 0.000 description 3
- 230000001795 light effect Effects 0.000 description 3
- 239000003550 marker Substances 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 210000001747 pupil Anatomy 0.000 description 3
- 238000009877 rendering Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 238000003786 synthesis reaction Methods 0.000 description 3
- 238000000844 transformation Methods 0.000 description 3
- 241000234282 Allium Species 0.000 description 2
- 235000002732 Allium cepa var. cepa Nutrition 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 2
- 238000010420 art technique Methods 0.000 description 2
- 210000005252 bulbus oculi Anatomy 0.000 description 2
- 239000003518 caustics Substances 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000001427 coherent effect Effects 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 210000005069 ears Anatomy 0.000 description 2
- 210000004709 eyebrow Anatomy 0.000 description 2
- 230000008921 facial expression Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000002715 modification method Methods 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 230000009897 systematic effect Effects 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 206010047571 Visual impairment Diseases 0.000 description 1
- 230000001594 aberrant effect Effects 0.000 description 1
- 230000003667 anti-reflective effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000009189 diving Effects 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 239000010437 gem Substances 0.000 description 1
- 229910001751 gemstone Inorganic materials 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000004886 head movement Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 239000001205 polyphosphate Substances 0.000 description 1
- 230000001179 pupillary effect Effects 0.000 description 1
- 238000007493 shaping process Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 239000002966 varnish Substances 0.000 description 1
- 208000029257 vision disease Diseases 0.000 description 1
- 230000004393 visual impairment Effects 0.000 description 1
- 239000011800 void material Substances 0.000 description 1
- 235000012431 wafers Nutrition 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/77—Retouching; Inpainting; Scratch removal
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/006—Mixed reality
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2219/00—Indexing scheme for manipulating 3D models or images for computer graphics
- G06T2219/20—Indexing scheme for editing of 3D models
- G06T2219/2012—Colour editing, changing, or manipulating; Use of colour codes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2219/00—Indexing scheme for manipulating 3D models or images for computer graphics
- G06T2219/20—Indexing scheme for editing of 3D models
- G06T2219/2021—Shape modification
Definitions
- the field of the invention is that of image processing and image synthesis.
- the invention relates to a method of occultation of an object in an image or a video.
- the invention finds particular applications in the field of augmented reality for fitting a virtual pair of glasses by an individual wearing during the fitting a real pair of glasses.
- This technique reconstructs the face statistically from images of individuals at an identical angle of view, generally from the front.
- This technique only works in two dimensions, considering only the inside of the 2D envelope of the face on the image. In other words, all the elements of the pair of superimposed glasses with a background of the face are not considered by this technique, which is penalizing on images with pairs of glasses wider than the face or when the face is not face on the image.
- Another major disadvantage of this technique is that it takes into account only pairs of glasses including a very thin frame, thus excluding all pairs of glasses having a thick frame.
- none of the existing augmented reality systems can virtually remove a real object, such as a pair of glasses, on one or more individuals carrying this object.
- the present invention aims to remedy all or part of the disadvantages of the state of the art mentioned above.
- One of the main objectives of the invention is to propose a technique that allows a user wearing a real-vision device to see himself on a screen, as in a mirror, without the actual vision device on the face and to try a virtual object replacing on the screen the actual vision device preserved on the face.
- Another objective of the invention is to propose a technique that is realistic for the user.
- Another objective of the invention is to propose a technique that operates in real time.
- Another object of the invention is to provide a technique that allows the user trying a virtual object to move the head in any direction.
- An object of the invention is also to propose a technique for removing the visible part of an object, in particular a pair of glasses, on an image or a video, but also local light interactions such as the reflections of the glasses or the shadows. worn.
- the object may be for example a vision device worn on the face, such as a pair of glasses or a portable device on the head comprising a mount and a display screen, such as a virtual reality helmet, mixed reality or augmented reality.
- a vision device worn on the face such as a pair of glasses or a portable device on the head comprising a mount and a display screen, such as a virtual reality helmet, mixed reality or augmented reality.
- the object can also be any other accessory worn on the head of an individual such as a scarf, a hat, make-up, a jewel or a hairstyle.
- the image is acquired by an image acquisition device that can be a camera, a camera or a depth camera.
- the depth camera well known to those skilled in the art, combines a camera and an infrared measurement of the distance of the elements relative to the objective.
- the image can be alone or included in a sequence of images, also called video.
- the method of generating an image comprises the following steps:
- the method makes it possible to modify the visual appearance of the detected object by covering it with a mask whose appearance is modified.
- the mask comprises pixels covering a continuous area or not on the initial image.
- the mask can cover all or only part of the object.
- the mask can cover only the frame of the pair of glasses, the frame and part of the glasses, the frame and the glasses in total, or only the glasses, it It should be noted that the shadows of glasses can also be covered by the mask,
- Changing the appearance of the mask corresponds to a change in the color and / or opacity of part or all of the pixels of the mask.
- the modification of the appearance of the mask comprises a step of replacing the texture of part or all of the object on the final image.
- the texture of the object is a representation of the external appearance of the object.
- the texture may for example be related to the color of the object, to its constitution, as the presence of different layers of porous or translucent materials.
- the texture may also be related to the type of coating of the object, such as the presence of a layer of gloss or matte varnish.
- the modification of the appearance of the mask comprises a step of determining the texture of part or all of the object, the texture reproducing the elements behind the plane of the object in order to conceal all or part of the object on the final image.
- the object detected in the initial image is automatically hidden from the final image.
- the method of generating a final image from an initial image is a method of obscuring an object in an image.
- the mask also covers all or part of the drop shadow of the object.
- changing the appearance of the mask also makes it possible to hide the shadows of the object.
- the shadow of a pair of glasses worn on the face of an individual can also be erased on the face, increasing the realism of the occultation of the pair of glasses.
- the method for generating an image also comprises the following step:
- the elements included in the second layer are for example hair covering a branch of a pair of glasses, a partially located hand in front of the object.
- the superposition of the different layers makes it possible to preserve the realism of the final image.
- the method of generating an image also comprises, before step b), the following steps:
- the initial image acquisition device comprises a photographic sensor and a photographic lens for converging real images on the sensitive surface of the photographic sensor.
- the photographic lens includes at least one converging lens.
- the image acquisition device can be for example a camera, a camera or a webcam.
- the orientation of the object with respect to the image acquisition device corresponds to the angles formed by the object in a reference of the acquisition device.
- This marker may for example be an orthonormal marker whose axis coincides with the optical axis of the lens. In other words, the object whose orientation is determined, is followed during a sequence of images.
- the characteristic dimension of the object may be for example the width of the frame of a pair of glasses.
- the method of generating an image also comprises, before step b), the following steps:
- the model representing the object is superimposed virtually on the object.
- the model of the object may include two-dimensional images that are distorted and flattened according to the orientation and size of the real object.
- the model of the object can also be three-dimensional with or without thickness.
- the orientation and the characteristic dimension of the model correspond to parameters of similarity between the model of the object and the real object.
- the projection of the three-dimensional model makes it possible to obtain the mask.
- the mask can cover the all or part of the resultant of the projection of the model on the layer. It should be emphasized that the mask can also cover an area of the image larger than the projection.
- the development of the model of the object is performed from at least one image of the object alone.
- the generation of the model of the object may for example be performed in a device dedicated to modeling, comprising a box in which the object is housed, and one or more image acquisition devices oriented towards the object.
- An image may be sufficient for the development of the model of the object, provided that it is a three-quarter view of an object with a plane of symmetry, such as a pair of glasses.
- the development of a model of the object is made from at least two images of the object, the images presenting the object from different angles.
- the object is worn on the face of an individual.
- the development of the model of the object is carried out from at least one image of the object worn on the face of the individual.
- the individual can keep the object on his face during the generation of the model.
- the object comprises a frame extending on either side of the face, and at least one glass assembled to said frame.
- the object may be a vision device such as a pair of glasses.
- the method for generating an image also comprises a step of identifying the frame among the frames previously modeled and stored in a database, the mask being developed from the model of the mount identified.
- the projection on the first layer of the model of the frame identified and previously modeled makes it possible to obtain a realistic mask of the frame.
- the mask may include all or part of the projection of the frame on the first layer.
- An area of the image corresponding to a glass assembled to the frame, or to a drop shadow, may be added to the mask.
- a database can store mount models and the mount is identified among the mounts stored in the database.
- the identification can be done automatically by the process or manually by an individual.
- the manual identification can be made for example by the information entered by the manufacturer inside the frame of the pair of glasses.
- the identification of the frame is performed by generating support curves that fit on the contours of the frame.
- the identification of the frame is based on at least one of the following criteria:
- the method for generating an image also comprises a step of producing a representation of the environment of the object.
- the environment includes all the elements surrounding the object on the image, as well as the elements in the background of the object on the image.
- the representation may be in the form of an image and / or a three-dimensional model.
- the representation of the environment may comprise a model of the face on which is worn the pair of glasses and / or an image corresponding to the background of the face.
- the step of modifying the appearance of the mask comprises the following substeps:
- the geometric projection of the representation of the environment on the intermediate layer makes it possible to obtain an image on which the mask of the object is superimposed.
- the geometric projection of the three-dimensional model on the intermediate layer produces an image superimposed on the background image.
- the intermediate layer thus presents a two-dimensional representation of the environment on which the mask of the object is superimposed.
- the three-dimensional model is textured.
- the determination of the new color implements an image editing method of the "Poisson image editing" type.
- the texture applied to the mask is adjusted colorimetrically according to the initial image in order to be confused with it.
- the edges of the mask are no longer visible in the final image and the mask can no longer be distinguished in this final image.
- This image editing technique described for example in the "Poisson Image Editing" document by P. Perez and M. Gangnet, notably solves a Poisson equation.
- the method of generating an image also comprises a step of detecting the presence of a face in the environment and in that the representation of the environment comprises a model of the detected face on which is applied a texture of the face.
- the texture of the face is a two-dimensional image that applies to the model. It should be emphasized that the model and texture can be advantageously realistic.
- the detection of the presence of the face can be done by detecting characteristic points of the face, such as for example the edge of the temples, the end of the nose or the chin, even the corners of the eyes.
- the method of generating an image also comprises a step of determining the orientation of the face with respect to the acquisition device and that the model of the face is disposed substantially according to the previously determined orientation.
- the three-dimensional model representing the face is realistically oriented in the virtual space corresponding to the scene acquired by the image.
- the mask at least partially covering the object worn on the face is developed from the geometric projection of the face model on the first layer.
- the occultation of the object worn on the face is achieved through a mask developed from a projection of the face model and not from a projection of the model of the object.
- this embodiment makes it possible to overcome the tracking of the object.
- the elaborate mask may not take into account the size of the object, in which case the size of the mask is set according to the size of the face. In the case of a pair of glasses worn on the face, the size of the mask is advantageously sufficient to cover most of the existing pairs of glasses.
- the method for generating an image also comprises the following steps:
- the model of the face is illuminated realistically compared to the real scene.
- the method for generating an image also comprises the following steps:
- the estimation of the colorimetric transformation can be performed by an analysis of at least one light source illuminating the face of the individual.
- the color of a pixel on the texture of the face is determined by means of an inpainting method from the colors of a patch near the pixel.
- the patch corresponds to a plurality of pixels forming a continuous zone.
- the shape of the patch may be square or rectangular, each side generally comprising between one and five pixels.
- a circular shaped patch can be obtained by inserting a Gaussian filter inside a square patch.
- the inpainting method well known to those skilled in the art, makes it possible to complete the texture of the face, especially in the case of the generation of the model of the face of an individual wearing a pair of glasses. Indeed, in this example, the frame or even the glasses mask part of the face.
- the position of the patch is located substantially on the perpendicular to the contour of the area comprising the missing pixels.
- the color of a missing pixel of the texture of the face is restored from a patch near the missing pixel, the patch being located on a perpendicular to the outline of the hidden area of the face.
- the position of the patch is located substantially vertically with respect to said pixel.
- the method of inpainting respects the general typology of a face which includes on both sides a vertical hair zone coming to cover part of the temples.
- the color of a pixel on the texture of the face is determined by means of an inpainting method based on the previously established and oriented model of the face, the model of the face including a representation of the eyes.
- the method of generating an image also comprises a step of identifying at least one eye zone on the texture of the face, the ocular zone corresponding to the position of an eye of the detected face.
- the identification of an ocular area on the texture of the face can be performed by identifying the position of characteristic points of an eye such as the precise outer and inner corners of an eye.
- the filling of the ocular area is performed knowing the topology of the detected facial eye.
- the topology of the eye includes a parametric representation through curves, different areas of the eye, including the iris and eyelids.
- the filling of the ocular area is more realistic because it respects the position of the iris and the pupil.
- the filling of the iris can be carried out by an inpainting method coming to recover a pixel nearby in a zone corresponding to the iris.
- the zone corresponding to the iris covers empty pixels or having no coherent values
- the iris is restored according to a standard iris topology possibly taking into account the color of the iris of the iris. other eye detected.
- the development of the representation of the environment of the object worn on the face of an individual is performed without detecting a face in the environment.
- the method is used without detecting or tracking the face of an individual.
- the elaboration of the representation of the environment comprises a sub-step of correcting the optical distortion due to a transparent element placed between the environment and an acquisition device. initial image.
- the method of generating an image is applied to all or part of a sequence of images forming a video.
- the video may be in the form of a recording or a stream in real time, such as a streamed video, a technique well known per se.
- the video can also be a real-time stream from a camera and visible on a screen instantly
- the representation of the environment and / or the model of the object are updated with each image of the sequence.
- the representation and / or the model being updated from several images of the sequence are more and more representative of reality.
- An area masked by the object such as the part of the face behind a pair of glasses, can thus be updated in the representation of the environment comprising a model of the face, when the individual turns his head.
- the device for acquiring the initial image takes images of the face under new angles of view, which improves the knowledge of the face.
- the representation of the environment and / or the model of the object is updated from a plurality of initial images taken according to a plurality of distinct angles of view. .
- the initial images taken at a plurality of distinct viewing angles may be from one or more image acquisition devices oriented at different angles.
- the generation of the final image is carried out in real time from the initial image.
- the processing of the acquired image is done in a weak and guaranteed time.
- the processing time of an acquired image of the individual makes it possible in particular to display the image of the treated individual without visible offset for the individual.
- the treatment time is less than 1 / 10th of a second.
- the treatment time is preferably but not necessarily less than the display time between two images, which is usually equal to 1 / 25th of a second.
- the real-time processing makes it possible to display a video stream coming from a camera instantly on a screen, the images of this stream having been processed in a sufficiently short time not to be perceived by the camera. human eye.
- the invention also relates to an augmented reality method for use by an individual wearing a portable device on the face, comprising the steps of:
- the portable device generally comprises a mount adapted to be worn on the head.
- the portable device may also include at least one glass and / or at least one screen.
- the frame can advantageously be configured to avoid covering the eyes and eyebrows.
- the portable device may be a vision device. This augmented reality process makes it possible for an individual to wear a pair of corrective glasses to see themselves on the screen with the same pair of glasses but with a different color and / or texture.
- the portable device is totally or partially hidden from the video displayed in real time.
- the portable device is a vision device comprising a frame and corrective lenses adapted to the vision of the individual wearing the portable device on his face.
- the portable device worn by the individual includes corrective lenses adapted to the view of the individual.
- the individual carrying the portable device tries a virtual object superimposed at least partially in the video on the portable device partially or completely obscured.
- the augmented reality method comprises a step of initializing the model of the face of the individual from at least one image of the individual not wearing the portable device on the face.
- the individual previously removed his portable device from his face for the generation of the model of the face, and delivers it after a given time.
- the acquisition of the image of the individual can be performed with one or more image acquisition devices.
- the individual can perform head movements so that facial model generation is performed from a plurality of facial images acquired from different angles of view.
- the augmented reality method comprises a step of initializing the model of the individual's face from a plurality of images of the individual wearing the vision device, the images corresponding to different angles of view of the face.
- the generation of the model of the face is performed without the user needing to remove his portable device from the face.
- the augmented reality method comprises a step of initializing the model of the portable device from at least one image of said device acquired in a dedicated modeling device.
- the augmented reality method comprises a step of initializing the portable device model from at least one image of the individual carrying the portable device.
- the invention also relates to an augmented reality device for fitting a virtual object by an individual wearing a vision device, the virtual object at least partially covering the vision device, the fitting device comprising:
- At least one camera acquiring a video of the individual
- a video processing unit acquired, the processing unit obscuring at least partially on the majority or all of the images of the video the viewing device via an image generating method
- At least one screen displaying the treated video of the individual.
- the screen is vertical and the camera is fixed substantially in the plane of the screen.
- this particular configuration allows an individual sitting or standing in front of the camera to see themselves live on the screen, as in a mirror.
- the device for fitting a virtual object comprises two cameras spaced, parallel to an edge of the screen, a distance of between thirty and fifty centimeters.
- the individual being generally placed at a distance of between eighty centimeters and one meter of the screen in order to be able to touch the screen, the distance between the cameras is optimal to obtain two shots of the face to reconstruct the model and the texture of the face in a realistic way.
- the fitting device of a virtual object further comprises a third camera substantially on the median axis between the first two cameras.
- the third camera makes it possible to obtain a front image of the individual, this image appearing on the screen.
- the first two cameras improve the realistic modeling of the face and the pair of glasses worn by the user.
- the screen is tactile.
- the virtual object may be a vision device such as a pair of glasses or sunglasses, a facial accessory or a makeup.
- the display of the acquired and modified video is performed in real time.
- the device is an augmented reality device where the user can try a virtual object and see themselves in real time on the screen.
- the device for fitting a virtual object comprises a device for acquiring the three-dimensional model of the vision device.
- FIG. 1 shows an embodiment of an augmented reality device for fitting a virtual object by an individual wearing a real vision device
- FIG. 2 represents a block diagram of an embodiment of a method for generating a final image from an initial image
- FIG. 3 represents in the form of a block diagram the steps of the method of generating an image with reference to FIG. 2;
- o 4a a perspective view of the model of the eyes
- o 4b a side view of an eyeball model
- o 4c a front view of an eye model
- o 4d a side view of an eyeball model including eyelid curves
- o 4th a front view of an eye model illustrating the movement of the iris
- FIG. 5 illustrates an example of acquired texture of a real face
- FIG. 6 illustrates the masks generated during the method of generating an image with reference to FIG. 2
- FIG. 7 represents another embodiment of an augmented reality device allowing the fitting of a virtual object by an individual wearing a real-vision device
- FIG. 8 represents in the form of a block diagram another embodiment of a method for generating a final image from an initial image
- FIG. 9 represents another embodiment of an augmented reality device used by an individual wearing a real-vision device
- FIG. 10 shows a pair of spectacles used in the exemplary embodiments of the invention.
- FIG. 1 1 shows a screen showing a video of an individual wearing a pair of glasses on the face
- Figure 12 shows a screen displaying the video with reference to Figure 1 1 in which the pair of glasses is obscured.
- FIG. 10 represents a pair of spectacles 1 1 1 comprising a rigid frame 1 12 and two corrective lenses 1 13 assembled to the frame 1 12.
- the frame 1 12 symmetrical with respect to the medial plane AA comprises a face 1 12b and two branches 1 17 extending laterally on either side of the face 1 12b.
- the face 1 12b comprises two circles 1 14 surrounding the lenses 1 13, a bridge 1 15 ensuring the spacing between the circles 1 14 and two pins 1 19.
- Two plates 1 16 each attached to a circle 1 14 are intended to rest on both sides of an individual's nose.
- the two branches 1 17 are each fixed to a pin January 19 of the face 1 12b by means of hinges January 18 thus allowing the mount 1 12 to be articulated.
- the face 1 12b is inclined at an angle of between 5 ° and 10 ° relative to the plane perpendiculareau plane formed by the axes of the branches 1 17. This angle is generally confused with the angle pantoscopic of the pair of glasses 1 1 1, that is to say at the angle of the face 1 12b with the vertical when the pair of glasses 1 1 1 is placed on the nose of an individual looking away without tilting the head , the plane of the branches being horizontal.
- Each branch 1 17 ends with a sleeve 1 17b intended to rest on an ear of an individual.
- the frame 1 12 thus extends laterally on either side of the face of an individual wearing the pair of spectacles 1 1 1.
- pair of spectacles 1 1 1 used in the following two examples of embodiment of the invention is a non-limiting example of real object erased from an image or a sequence of images by the method which is the subject of the invention.
- Figure 1 shows a device 100 for fitting a virtual object 1 10 by an individual 120 wearing the pair of glasses 1 1 1 on the face.
- the individual 120 is moderately myopic.
- the visibility of the individual 120 not wearing a pair of corrective glasses is about twenty centimeters.
- the device 100 comprises a touch screen 130 fixed vertically on a support 131, a camera 132 centered above the screen 130, two peripheral cameras 133 and a processing unit 134.
- the device 100 further comprises a depth sensor measuring by infrared the distance of the elements relative to the camera.
- the depth sensor may include an infrared projector and a photosensitive sensor in the infrared wavelength.
- the photosensitive sensor being in the immediate vicinity of the projector, the density of the points of the image makes it possible to deduce therefrom a depth map indicating the distance of each point of the image relative to the sensor.
- the device 100 also comprises a scanner or a dual sensor for acquiring a model of the entire face of the individual 120.
- the individual 120 When the individual 120 is facing the screen 130, the individual 120 sees the image of his face 121 from the front, acquired in real time by the camera 132. In order to be able to touch the touch screen 130, the 120 individual stands at a distance of about one arm of the screen 130. The distance between the individual 1 20 and the touch screen 130 is between sixty and one hundred and twenty centimeters. The individual 120 wears the pair of glasses 1 1 1 in order to see the screen 130 clearly.
- the two peripheral cameras 133 are fixed on a parallel rail 135 to the upper edge of the screen 130, symmetrically on either side of the camera 132.
- the distance between the two peripheral cameras 133 is between thirty and fifty centimeters.
- the two cameras 133 are spaced forty centimeters apart from each other, which makes it possible to obtain images of the face 121 of the individual 120 with an angle of view shifted by about 20 ° by compared to normal.
- the processing unit 134 generates from each initial image of the sequence acquired by the camera 132 a final image of the face 121 of the individual 120 in which the pair of glasses 1 1 1 is obscured. In other words, the pair of glasses 1 1 1 is made invisible on the real-time display of the face 121 on the screen 130.
- a virtual representation of the scene acquired by the camera 132 is created.
- This virtual representation comprises a three-dimensional model of the pair of glasses 1 1 1 positioned on a representation of the environment comprising a model of the face of the individual 120.
- the projection of the model of the pair of glasses 1 1 1 and the representation of the environment allows to create a mask superimposed on the actual pair of glasses on each image of the sequence acquired by the camera 132.
- a virtual camera replaces the camera 132 with the same angle of view and the same magnification.
- the optical characteristics of the virtual camera are identical to those of the camera 132.
- the processing unit 134 thus generates a new image 210 from each image 220 of the sequence 200 acquired by the camera 132 according to a method of generation 300 of an image.
- FIG. 3 represents in the form of a block diagram the generation method 300.
- the generation method 300 detects the presence of the pair of glasses 11 1 in the initial image 220.
- the generation method 300 determines in a second step 320 the orientation of the pair of glasses 1 1 1 with respect to the camera 132.
- the generation method 300 determines in a step 330 a characteristic dimension of the pair of glasses January 1 on the initial image 220.
- the characteristic dimension is in the present non-limiting example of the invention, equal to the width of the frame 1 12.
- the generation method 300 produces in a step 340 a three-dimensional model of the pair of glasses 11 1 in a virtual space representing the real space acquired by the camera 132.
- the step 340 of developing the model of the pair of spectacles 1 1 1 comprises a first substep 341 for identifying the pair of spectacles 1 1 1 among the pairs of spectacles previously modeled and stored in a connected database. to the processing unit 134. This identification can be made by knowing the reference of the telescope and so-called framemarking elements printed on the latter.
- the identification of the pair of glasses 1 1 1 can also be performed by an automatic recognition from images of the pair of glasses worn by the user or in a device dedicated to the acquisition of images of the pair of glasses. glasses alone, such as a box of light.
- the automatic identification uses methods of indexing and visual recognition of the appearance of 3D objects well known to those skilled in the art, for example by generating support curves that adjust to the contours of the pair of glasses 1 1 1.
- the visual recognition of the pair of glasses can be made from the criteria of:
- the model of the pair of glasses 1 1 1 is extracted from the database during a substep 342.
- the 3D model of the pair of glasses 1 1 1 is developed, during a substep 343, to from images of the sequence 200 acquired by the camera 132, and possibly from the parameters representing the closest fit model determined during the search step in the base
- the images of the sequence 200 present the individual 120 wearing the pair of glasses 1 1 1 on the face.
- the development of the model of the pair of glasses 1 1 1 is thus carried out in real time from the images acquired by the central camera 132 and by the peripheral cameras 133.
- cameras acquire images from a new angle of view.
- the model of the pair of glasses 1 1 1 is updated with each image, in particular when the image presents a view of the individual 120 under a different angle.
- the eyeglass pair model 1 1 1 developed during the sub-step 343 is constructed by first creating a shape pattern of the face 1 12b of the pair of glasses 1 1 1 and a model of the branches 1 17 of the pair of glasses 1 1 1. It should be emphasized that in the case where the pair of glasses is not symmetrical, a model for each branch is created.
- the shape model of the face 1 12b also includes the wafers 1 16.
- a skeleton of the pair of glasses is used.
- the skeleton is extracted from a database containing typical topologies of pairs of glasses.
- the typical topologies of pairs of glasses make it possible to classify the pairs of glasses according to the shapes of the glasses. Topologies are defined by:
- a bridge or a bar connecting the two glasses, the branch and / or the bar can be single or multiple;
- the thickness is determined around the skeleton of the pair of spectacles by generating a closed 3D envelope that encompasses the pair of spectacles 1 1 1.
- the generation of the 3D envelope is done in three sub-steps: creation of support curves in planes perpendicular to the skeleton. These support curves substantially correspond to the sections of the frame January 12;
- the support curves for generating the 3D envelope are derived from a priori knowledge, drawn manually or learned statistically.
- the initialization of the support curves is generally performed during the visual recognition step in an attempt to automatically identify the pair of glasses 1 1 1.
- the support curves are generated from the images of the pair of spectacles 1 1 1 worn on the face or from images of the pair of spectacles 1 1 1 acquired on a neutral background by a dedicated modeling device (not shown in Figure 1).
- the model of the pair of glasses 1 1 1 1 is then recaled identically to the actual pair of glasses 1 1 1, during step 350.
- the model of the pair of glasses 1 1 1 and has the same orientation relative to the camera 132 and the same characteristic dimension to the image that the pair of glasses 1 1 1 real.
- the model of the pair of glasses 1 1 1 is positioned in the virtual space, oriented according to the position of the virtual camera and configured according to the size of the actual pair of glasses 1 1 1.
- a magnification factor can thus be applied to the model of the pair of glasses 1 1 1.
- the fitting parameters of the model of the pair of glasses 1 1 1 are noted Pe Mg .
- the generation method develops a geometric model M a in three dimensions of an avatar representing the face without the pair of spectacles 1 1 1.
- a texture T aNG of the face without the pair of glasses 1 1 1 is also created during the step 355.
- the geometrical model M a is parameterized in morphology and expressions according to the method of elaboration of the model of the face described in the following.
- the method of developing the avatar includes a first step of detecting the face in the image and facial analysis of the detected face. The detection of the face is carried out in the present non-limiting example of the invention by a Viola-Jones method, as explained in patent FR2955409.
- a line alignment algorithm is then used to find specific facial features during a second substep of the facial shaping process.
- a line detector well known to those skilled in the art is used and can very reliably find internal features of the face.
- HPAAM line alignment algorithm described in the European patent application EP2678804, then makes it possible to locate precisely the projection of significant 3D lines on the image. Unlike existing techniques that result in location error in congested environments, HPAAM is particularly stable on features on the contour of the face, such as points of the ears. Since the HPAAM algorithm is a learning phase technique, the use of predetermined points having a 3D match has an impact on the success of the overall facial analysis technique, particularly with respect to robustness and efficiency of the technique. Typically, this relationship is specified for a small number of points in 3D facial analysis techniques, such as the starting points of a 3DMM fitting strategy, in which five points are set manually.
- this step of detection of the face is robust to occultations of the face constituted by the pair of glasses 1 1 1.
- the second step of the process of developing the avatar concerns the estimation of the parameters of the model of the face 0 model comprising:
- the extrinsic parameters Pe Ma of the face model that is to say the parameters for laying the face, including the position and the orientation of the face
- the intrinsic parameters Pi Ma of the face that is to say the 3D morphology of the face
- the parameters of the model 0 face model are estimated using a statistical geometric model of the morphology of the human face.
- a database of faces is used, such as for example the database described in the Blanz and Vetter document published in 2003, entitled "Face Recognition Based on Fitting a 3D Morphable Model”.
- An estimate of the parameters of the model 0 face model and cam 9 parameters of the virtual camera is performed using the features found at the line detection stage and dynamically adjusting contours in the image.
- the function Proj) represents the projective transformation of a 3D scene, for example the face or the pair of glasses, towards a layer or image plane, by considering a pinhole-type camera model, well known to those skilled in the art. , which allows to perform a perspective division.
- this projection ProjCX; K, R, T It should be emphasized that the projection of the parametric face contours corresponds to the projection of points of the face model whose normal is orthogonal to their direction of observation.
- a sampling of the direction orthogonal to the contours of the current sample point is performed and allows sampling of the contours for several reasons: numerical efficiency, stability and compatibility with other alignment techniques used in the pursuit of objects 3D.
- a calculation of C e cam , e model ) which contains a subset of points of the face model Xj with the normal n ; orthogonal to the axial projection and associated points of the contour of the image contj ProjContour ⁇ Xj, nj) is performed, where ProjContour is a function projecting the point Xj and seeking, along the normal projection, the best contour among multiple hypotheses.
- the cost function can for example be implemented using an estimator type M approach such as that using Tukey's robust weight function.
- a calculation of the residual for the nearest point along the normal direction of the contour among multiple assumptions can be made.
- the main advantage of this estimation technique is that when multiple images are available, as here in the image sequence 200, it extends to a multi-image analysis algorithm that relaxes the 2D semantic correspondence constraint. / 3D and allows to refine all estimated parameters. It can be used to search for the morphology of best fit for all images.
- the third step in the avatar development process involves adding 3D facial expressions.
- the expressions add a certain variability to the facial model and their exclusion allows a more stable and accurate estimation of the pose parameters and morphology of the face.
- One approach usually used to create parametric variations of a mesh is to use blend shapes, that is, a set of linearly combined geometric models to produce unique instances.
- blend shapes that is, a set of linearly combined geometric models to produce unique instances.
- a technique commonly used to calculate these forms of mixing is to deduce them statistically as described in [A 3D Face Model for Pose and Illumination Invariant Face Recognition, Paysan et al., 2009].
- the model has the following form:
- g m is the average 3D shape
- a is a vector which contains the adaptation parameters specific to the user
- V is a matrix which contains the basic forms Statistics.
- the bases of Statistical Forms only include the variations of identity, without taking into account variations of expression, so as to guarantee good aptitude for the separation of the control parameters.
- the model of the face includes an eye model connected by a rigid translation T SE between the reference of the model of the face and the reference of the model of the eyes.
- the two centers of rotation 402 are connected to the reference system of the eye system SE, by a distance pdS, SG ⁇ R, L ⁇ .
- Each eye 401 is oriented relative to the reference system of the eyes by angles rxe, ryeS, SG ⁇ R, L ⁇ , respective rotations about the x and y axes.
- the centers of rotation 402 are at a distance dr from the center of a disk 403 of radius hdi representing the iris.
- Disk 403 is comprised in an element composed of three 3-position Bezier curves 410 having the same start and end control points, pEo, pEi, as shown in Figure 4c.
- the curves of the edges of the eyes can be represented in 3D on the mesh of the face. It is important that the points pEo, pEi are at the intersection of the curves of the edges of the eyes, and that the curve which moves is parameterized by a parameter dpELv allowing the curve eyelid 41 0 3 to evolve between the values of the high curve 41 0i and the low curve 410 2 .
- This one-dimensional parameter can influence the 3D path of the curve 41 0 3 of the eyelid in a curve defined in space.
- curves 41 0i and 41 0 2 are controlled by control points comprising respectively the points pEuL and pEuR, and the points pEdL and pEdR.
- the 3D path of the curve 410 3 of the eyelid can be represented in the deformation modes of the parameterizable morphology model, as a function of the displacement of the position of the eyelid dpELyEt) given by the rotation rxEL around the x axis of the reference mark particular of the eye, where t between 0 and 1 makes it possible to parameterize the position of a point on the eyelid curve 41 0 3 .
- the T SE relationship allows the iris discs to rotate around the PL and PR points to touch the eyelid curves.
- the difference between the sampled points and the contour image can also be determined by a search according to the method of normals previously described.
- K is the matrix of the intrinsic parameters of the camera
- Itfprojr m , SE is the image generated by the projection of the avatar model and the eye system taking into account occlusions of the eye system SE by closing the eyelids or by self-occultations due to the pose of the model.
- the generation of the image assumes a known texture.
- a resolution is added on learning parameters that vary the texture parametrically, of the active model type of appearance.
- the contour difference is advantageously used for initialization for reasons of performance and simplicity of the data.
- the set of points of the image CQ is selected along the normal to the gradient at the considered curve C l or C E projected, for the points of the curves associated with the values of the parameters ⁇ and s.
- the ProjContour function is also used for the minimization of the eyes. It should be noted that in order to make the system of equations robust at first initialization, the following default values are used:
- the estimation of the extrinsic and intrinsic parameters of the face is improved. Indeed, these values are used to perfect the estimation of the parametric model. If the parametric model does not fully correspond because the parameterization does not explain the depth, the face model is adapted to the surface, solving the system described in equation (1) 3D face resolution. We then have not an estimate of the parameters of the face but a metrological model parameterized the face of the user.
- step 356 the textures of the face T aNG and the background T bg , defined in more detail below, are updated. during step 356 to correspond to the reality of the scene acquired by the camera 132.
- the texture T aNG 450 illustrated in FIG. 5, is an atlas of the face, calculated according to the conventional mesh unfolding methods well known to those skilled in the art. Once the 3D face is projected in the image, the faces visible and oriented towards the camera, for example by z-buffer or culling methods, make it possible to fill the texture image T aNG 450.
- the textures of the eyes are distributed over the texture of the face and are broken down into three parts: the texture of the iris T aNG l 451, the texture of the white of the eye Ta NG E 452, the texture of the eyelid T aNG EL .
- These three elements can be incomplete at the time of acquisition but can be completed in a simple way by interpolation for T aNG E and T aNG EL for areas not known to be synthesized or by knowledge of the topology for the non-visible parts, such as the top of the iris if the eye is not wide open.
- the circular character of the pupil and the iris makes it possible to complete the texture according to a polar parameterization.
- a map of the bottom T bg also called a background or background map, is developed in step 357.
- the map T bg corresponds to the background and to all that is considered belonging neither to the real pair of glasses worn by the user, nor to the face, or any other element explicitly modeled, such as a model of hair or a hand coming from to be superimposed on the face and the pair of glasses 1 1 1.
- the map T bg is updated dynamically by following update rules such as found in conventional background subtraction techniques (in English "background subtraction"). Predominant color patterns are used for each of the pixels, using probability distributions and possible modes for the colors. Several models can be used, such as mixtures of Gaussians, or estimates of modes by methods of nuclei on histograms. This model is coupled to a dynamic model of temporal and possibly spatial update.
- the dynamic updating model can be done in the following way: as in [Active Attentional Sampling for Speed-up of Background Substraction, Chang et al., 2012], for each pixel, we take into account a property of temporality P t , a property of spatiality P s eliminating the isolated pixels and a property of frequency on the last images of the video P f making it possible to eliminate the pixels changing of class too often and possibly due to the noise. The product of these three values gives a probability for each pixel to belong to the map and to be updated.
- the background map T bg is initialized by all the pixels not considered as the projected face or the pair of glasses projected at this stage.
- the background map has the same dimensions as the image 220.
- the modification method includes the steps during which:
- a segmentation map of the face T a is calculated from the projection of the face model M a in the image.
- the projection of the glasses model M g makes it possible to obtain the segmentation map of the glasses T g .
- each pixel p is treated as follows:
- the occultation map is represented by a dynamic texture T fg that is updated with each image of the sequence 200.
- T fg dynamic texture
- the occultation map is associated with a geometric model M fg that can be variable. This can be a plane that represents a layer in front of the 3D scene, or an estimated or available depth map.
- the value of the occultation map is determined by difference of the appearance prediction with the real image, that is to say by the difference between the projection of the virtual models representing the face, the pair of glasses and the background. and the real image. In other words, the occultation map includes all the elements that have not been modeled before.
- an inpainting technique is used to fill any void spaces in the occultation card, thereby improving the appearance of the occultation card.
- the occultation map takes into account degrees of local opacity. This change in opacity is commonly used to solve digital matting problems.
- Ta fg the grayscale opacity channel of the occultation map
- TB fg its binarization for the opaque pixels of value 1.
- the detection of occultations is easier and methods well known to those skilled in the art can be applied.
- the RGBD sensors based on infrared technologies get a very bad signal because the pairs of glasses are objects usually made of complex and dynamic materials, like the metal, the translucent plastic and glass. The diffraction and refraction effects of these materials prevent the depth map creation system from working properly.
- the spatial resolution of these sensors is not sufficient for very thin glasses. As a result, not only are the glasses not or very poorly identified by the system, but they corrupt or render inaccessible all face data located in the neighborhood and behind.
- the use of the image and the parametric model of the pair of glasses proposed makes it possible to overcome these structural problems of the depth sensors.
- the textures of the pair of glasses, the face and / or the background are completed and updated during step 359 of the method of generating an image.
- the state of the cards representing the elements displayed in the image changes according to the knowledge of the elements.
- the face of the individual 120 is partially masked by the pair of glasses 1 1 1.
- New elements of the face of the individual 120 appear when the individual 120 turns his head.
- the color information may also be distorted because of the refraction of the glasses, especially in the case where the glasses are tinted or because of the shadows worn on the face of the pair of spectacles 1 1 1.
- Statistical learning models are used on the facial area, but are less effective on the background. It is thus possible to replace the pixels of the facial area with known techniques such as active models of appearance or 3D models morphables (in English “3D morphable models”).
- the spatial localization technique is used.
- the filling technique close to the inpainting techniques well known to those skilled in the art, relies on texture synthesis, providing relevant knowledge for the reliable and real-time resolution of the filling problem. Since the topology of eyeglass models is known, and the real-time constraint is important, the patch filling is used, which guarantees the continuity of the colors between the painted areas and the respect of the structures of the textures. This technique allows us to quickly search for similar elements in the image as well as parallel processing of the majority of the pixels to be replaced.
- the real-time filling technique is based on an inpainting technique well known to those skilled in the art.
- the areas to be treated are filled by pixel or patch, using a three-step algorithm:
- the patch is formed by a square window centered around a pixel.
- the knowledge of the topology of the face makes it possible to define the directions and priorities of the course and the sampling zones prior to the patches. For example, if the eyes are hidden, the geometric model of eye construction is parametrically known in advance, and the priority, the size of the patches and the direction of propagation can thus be adapted according to the related curves. to the particular topology of an eye.
- the propagation of structures is never far, whatever the pose of the face, the direction of propagation of the isophotes.
- the pair of glasses although may include a thick frame, has projections in the image such that the different edges of the same sub-object are almost parallel.
- the search for the patch is reduced to the first patch found containing information in that direction.
- the pixels replaced in the previous iteration are used, allowing continuity of the structure.
- the propagation direction is also predefined and is only calculated for patches judged to be highly structured by a structural criterion. For example, the entropy of the patch under consideration, or a coefficient dependent on the gradient direction standard, may be used. This approach avoids a systematic and expensive ranking of priorities as well as propagation directions.
- the patch fill ⁇ 0 centered around the pixel p0 at a distance from the front of the mask to be filled such that the patch contains pixels of a known area.
- d max a maximum confidence distance which guarantees continuity of the structures, and we move in both directions from the normal direction to the contour n c of the mask mask to find the two full patches T1 and T2 (centered in pT1 and pT2 ) in the nearest "texture" areas.
- This technique saves search calculations from the nearest colorimetric patch.
- the colorimetric adaptation is then performed to fill the pixels of the patch ⁇ 0, taking into account the distance d1 from p0 to pT1 and the distance d2 from p0 to pT2 in order to allow the following linear interpolation:
- each pi (u, v) T corresponds to a pixel of a patch T.
- the notation ssi corresponds to the abbreviation of "if and only if”.
- patch sizes are dependent on the size of the structure to be replaced, namely the thickness of the pair of glasses, and the distance from the user to the camera.
- Figure 6 illustrates the development of the masks from the image 220 of the user 120 ( Figure 6a). As illustrated in FIG. 6b, the environment i b9 in the background of the pair of spectacles 1 1 1 is decomposed into several zones:
- the area 470 may be subdivided into semantic subregions 472, corresponding for example to the region of the hair 472-, to the region of the skin 472 2 .
- the generation method elaborates the mask of the pair of spectacles 1 1 1 by geometric projection of the three-dimensional model of the pair of spectacles 1 1 1 on a first layer.
- the first layer is previously emptied before the geometric projection.
- the first layer comprises an image of the three-dimensional model of the pair of glasses 1 1 1 according to the same angle of view and the same size as the actual pair of glasses 1 1 1.
- the mask TM g of the pair of spectacles 1 1 1 is divided into several parts: the mask TM gf of the frame 1 12b and the branches 1 17; and
- the TM ge mask also includes the lighting effects on the lenses, especially the reflections.
- the mask TM gf corresponds to the RGBA rendering image of the model of the pair of spectacles 1 1 1 for the values of face parameters Pe Ma and glasses Pe Mg estimated at the instant corresponding to the image 220.
- the mask TM gf takes into account possible occultations of the pair of glasses 1 1 1 such as for example a hand placed in front of the face or a lock of hair falling on the face.
- a TMB gf binary mask is obtained by binarizing the rendering alpha layer of the TM gf mask. Since the alpha channel represents the transparency of the pixels, the binarization of the alpha layer delimits the TM gf mask.
- FIG. 6c represents the environment represented in FIG. 6b on which the TMB gf mask is added.
- the TM gl and TM ge masks are determined using the same technique as for the TM gf mask, respectively considering for each mask the lenses 1 13 and the light effects such as reflections on the lenses or shadows worn on the face.
- the mask TM a corresponding to the face is created during the step 365 from the model of the face comprising the eyes, according to the orientation and positioning parameters of the face estimated previously for the image 220.
- the TMB binary mask gf of the pair of spectacles is contained in the face region TM a or in the background map T bg , as can be seen in Figure 6c.
- a sampling is performed in a locality defined on either side of the TMB mask gf , according to a parameterization given by the topology of the pair of glasses, in the direction of the normal to the contour n c .
- the branch is sampled on each side on zones of maximum size representing a partition n R of the regions defined by TM a or T bg .
- an adjustment of the spatial division is made to the boundary curves of the regions. Thanks to this division, it is possible to estimate the field of local colorimetric transformations between the appearance prediction and the current image, for the face areas TM a and T bg background deprived of the glasses region TM g , which allows find transformations due to global light changes, or drop shadows
- areas not responding to this colorimetric dynamic may not be considered at first, such as eyebrows, hair or beard, to focus on the skin, which follows a pseudo-Lambertian dynamic and allows low and medium frequency matching of colors. These areas are identified and segmented by the points and curves found during feature recognition and can be refined in the texture space. The zones of the same type are then calculated on the transformation, as in the color transfer or tone mapping techniques well known to those skilled in the art.
- This field of colorimetric transformations is applied respectively to the images TM a and T bg to form the cards TM a Wc and T bg Wc. It should be emphasized that the colorimetric transformation is performed on colorimetrically consistent subregions of the TM a and T bg images. These coherent sub-regions can advantageously be included in a semantic sub-region 472 in order to improve the final result. In addition, the color transformation takes into account the dynamic differences between the subregions of these spaces.
- These new images TM a Wc and T bg Wc are used to analyze the pixels of the current image / whose color is not determined by the prediction, in particular in the glasses and face areas, in order to detect the reflections and light changes and geometric TM gh glasses as well as the shadows of the TM ge mount in the associated regions.
- This technique makes it possible in particular to correct the facial deformations due to the optical correction of the glasses of the pair of spectacles 1 1 1 worn by the user 120.
- the threshold e is large enough to encompass aliasing colors and avoid compression image artifacts and sensor.
- the mask can then be expanded according to the confidence in the knowledge of the 3D object and the registration.
- FIG. 6d represents the image represented in FIG. 6c on which is added the TMB ge map representing the light effects, reflections and shadows.
- the map of the pixels to be replaced TMB g is the union of TMB maps gh TMB ge , and TMB gf , deprived of the pixels of the occultation alpha map TB fg .
- TMB g U [TMB gi , TMBg e , TMB gf . ⁇ ⁇ TB fg
- the occultation alpha map TB fg represents the opaque pixels of the occultation map T fg , that is to say the pixels of T fg whose alpha value is equal to 1.
- FIG. 6e represents the image represented in FIG. 6d in which the occultation alpha map TB fg is added.
- the modification of the appearance of the mask representing the pair of glasses January 1 is carried out during the step 370.
- the modification of the appearance replaces the pixels of the image 220 corresponding to the TMB binary mask g by the appropriate values that make it possible to make disappear or apply a treatment on the target parts of the pair of glasses 1 1 1 in the image 220.
- the colors chosen may be from the following techniques or their combinations:
- the preferred technique in this example is the color replacement by prediction, because it best manages the discontinuities of the model. Although it may be sensitive to an estimation error, the addition of a mask expansion and a color continuity constraint make it possible to propose non-detectable replacement results for the human eye. Thanks to the calculated maps TM a Wc and T bg Wc, and the map T fg , all the pixels can be replaced in the majority of the cases.
- Another preferred technique, generally used, is an image editing technique known as "Poisson image editing". This technique consists in solving the color of the pixels of the zone to be replaced by guaranteeing a continuity to the contours of the mask. It should be emphasized that this technique changes the colorimetry while keeping the structure of the texture applied to the mask. This texture is indeed generally deformed, for example by projection to obtain a texture adapted to the environment of the pair of glasses 1 1 1.
- the final image 210 is then generated during step 380 of the method
- first layer comprising the mask of the pair of spectacles 1 1 1;
- the virtual pair of glasses 1 10 is positioned on the face 121 of the individual 120 through an intermediate layer inserted between the first layer and the second layer.
- the intermediate layer comprises a projection of a model of the virtual pair of glasses 1 10 realistically positioned on the face 121 of the individual 120.
- the skilled person can for example refer to the application FR 10 50305 or the application FR 15 51531 describing in detail the techniques for fitting a virtual pair of glasses by an individual.
- FIG. 7 represents a device 500 for fitting a virtual object 510 by an individual 520 carrying the pair of spectacles 11 1 on the face.
- the device 500 comprises a touch screen 530 fixed vertically on a support 531, a camera 532 centered above the screen 530, two peripheral cameras 533 and a processing unit 534.
- the device 500 also comprises a device 537 for measuring the distance of an element from the screen 530, comprising an infrared projector 535 projecting a pattern and an infrared camera 536.
- the device 500 further comprises a modeling device 540 comprising a turntable 541 intended to receive a pair of spectacles at its center, two digital cameras 542 fixed, oriented towards the center of the plate 541 and a plain background 543 intended to be backwards. of the pair of glasses modeled.
- the modeling device 540 connected to the processing unit 534 can thus actuate the turntable 541 and acquire images of the pair of spectacles January 1 at different angles of view.
- the plate of the modeling device 540 is fixed.
- the modeling device 540 then comprises two additional fixed digital cameras oriented towards the center of the plate.
- the position of the two additional cameras corresponds to the rotation of 90 degrees of the position of the two cameras 542 around the central axis normal to the plate.
- modeling device 540 performs a calibration by acquiring for each camera 542 an image of the plain background 543 alone.
- the individual 520 removes the pair of glasses 1 1 1 he wears on the face and the place, branches 17 open, in the center of the turntable 541. In order to properly position the pair of glasses 1 1 1, markers are provided on the plate 541.
- the first camera 542i oriented so that the optical axis of the camera 542 acquires a front image of the pair of glasses 1 1 1, then after a 90 ° rotation of the turntable 541 an image this side of the pair of glasses 1 1 1.
- the second camera 542 2 acquires images of the pair of glasses 1 1 1 for diving, 3/4 face and 3/4 back.
- the position of the camera 542 2 is thus raised at approximately 45 ° with respect to the median plane of the plate 541.
- a three-dimensional model of the pair of glasses 1 1 1 is created from the four acquired images of the pair of glasses 1 1 1 and two images of the bottom.
- the pair of glasses 1 1 1 is segmented in each acquired image by making the difference between the background images and the images with the pair of glasses 1 1 1, which allows to create binary masks of the different elements.
- the frame 1 12 of the pair of glasses 1 1 1 is considered as an assembly of three 3D surfaces:
- each branch 1 17 of the pair of spectacles 1 1 1 a surface for each branch 1 17 of the pair of spectacles 1 1 1. It should be noted that since the pair of spectacles 1 1 1 1 is symmetrical, the two branches 1 17 are similar and only the opening angle between each branch 1 17 and the face 1 12b can vary. Only a three-dimensional model of a branch 1 17 is thus generated. The three-dimensional model of the other branch 1 17 is then developed from the model of the first branch 1 17 symmetrically with respect to the main median plane of the first branch 1 17.
- a calculation of the distance map is performed for each of the images from the masks extracted from the segmentation.
- the estimation of the parameters of the 3D surface is done via a minimization respecting the criteria of central symmetry and continuity of the frame of the pair of glasses 1 1 1.
- An estimate of a 2D contour of the face 1 12b and the branches 1 17 is performed from the bit masks of the face 1 12b and branches 1 17.
- the 2D outline is then projected onto the corresponding 3D surface.
- a thickness is added to each of the 2D contours projected on the surfaces to obtain the three-dimensional models of the face 1 12b and the branches 1 17 forming the three-dimensional model of the pair of spectacles 1 1 1.
- a Delaunay triangulation is performed from the points of the 2D contours. This triangulation is used on the points of the 3D surface to create the model of the pair of glasses 1 1 1.
- the acquired images of the pair of glasses 1 1 1 are applied in texture on the model of the pair of glasses 1 1 1.
- 3D statistical models of each element of the pair of glasses 1 1 1 can be used for parameterizing and meshing 3D surfaces from 2D contours.
- An image of the individual 520 without glasses is acquired by the camera 532.
- a model M av of an avatar representing the individual 520 is developed from the acquired images and measurements of the distance to the screen of the elements of the image , following the process of developing the avatar model M described above in step 355 of the first exemplary embodiment.
- the device comprises three cameras, for example a camera centered vertically with respect to the screen and two cameras horizontally positioned symmetrically on either side of the central camera. These three cameras make it possible to obtain three images of the individual 520 with different angles of view in order to improve the representation of the face of the individual.
- a flattened texture of the individual 520's face is extracted from the avatar model M a .
- the pair of glasses 1 1 1 is followed in the sequence of images acquired by the camera 132 by a tracking method 600 of the pair of glasses 1 1 1.
- the tracking method 600 illustrated in the form of a block diagram in FIG. 8, comprises a first initialization step 610.
- the initialization step 610 makes it possible to position the model M g of the pair of spectacles 1 1 1 on the avatar M a and to open the branches of the model M g in the same manner as the pair of spectacles 1 1 1 actual placed on the face of the individual 520.
- a first positioning of the model M g is done in 3D on the avatar M a so that the model of the pair of glasses 1 1 1 rests on the nose and ears of the avatar.
- the model M g is thus positioned according to calculated laying parameters.
- the installation parameters include the orientation relative to the camera and the magnification to be applied to the model M g to obtain the pair of glasses 1 1 1 displayed in the image.
- the avatar is positioned and oriented according to the virtual camera having the same orientation and the same optical parameters as the camera 532.
- the position and the orientation of the face are determined on each image by means of a method of facial tracking well known to those skilled in the art.
- the tracking of the face is based on the tracking of characteristic points of the face.
- the characteristic points masked to the image especially those lying behind a pair of glasses or behind tinted lenses, are not taken into account in the tracking of the face.
- a projection on a first layer superimposed on the initial image, the model of the pair of glasses 1 1 1 positioned on the avatar provides a mask of the pair of glasses 1 1 1.
- the pose parameters are calculated by minimizing a cost function based on two components:
- the tracking method 600 selects, during a second step 620, the set ⁇ of the points of the model M g whose normal is substantially perpendicular to the axis formed between the point and the virtual camera.
- the tracking method 600 selects a subsample of n points from the set ⁇ of the points of the model M g .
- the subsample comprises a small number or no points of the branches.
- the vectors n2D m ln corresponding to the projections of the normals of the n points of the set ⁇ are calculated during the fourth step 640. From the projections p2D and the vectors n2D, the method 600 performs for each index m, a search of the point p_grad m of the image having the highest gradient along the projection p2D m of the normal at the point n2D m .
- the tracking method 600 then minimizes during the fifth step 650 the function of calculating the distance between the points p2D and p_grad.
- the position of the model M g is considered representative of the actual position of the pair of glasses 1 1 1.
- a mask covering the pair of glasses 1 1 1 is created from the projection of the model M g on the first layer.
- the modification of the appearance of the mask of the pair of glasses 1 1 1 is performed by replacing the color of the frame 1 12 of the pair of glasses 1 1 1 actually worn by the individual 520 by a new color.
- Regularization of the brightness is performed in order to make the modification of the color of the frame 1 12 realistic.
- the individual 520 sees his image on the screen 530 with the same pair of glasses 1 1 1 but including a different color of the frame 1 12.
- the modification of the appearance of the mask of the pair of spectacles 1 1 1 makes it possible to conceal the pair of spectacles 1 1 1 in the final image so that the individual 520 wearing the pair of glasses 1 1 1 on his face sees his picture on the screen 530 without the pair of glasses 1 1 1.
- FIG. 9 represents an augmented reality device 800 used by an individual 820 carrying the pair of glasses 11 1 on the face.
- the pair of glasses 1 1 1 is mounted corrective lenses adapted to the view of the individual 820.
- the individual 820 is facing a camera 832 connected to a screen 830 displaying the image of the head 821 of the individual 820 as in a mirror.
- the image displayed on the screen 830 shows the head of the individual 820 without the pair of glasses 1 1 1 on the face of the individual 820.
- the individual 820 can thus be seen clearly without his pair of glasses, as if he was wearing lenses.
- the sequence of images, also called video, displayed on the screen 830 a method of generating a final image from an initial image is used.
- the pair of glasses 1 1 1 1 is detected and tracked on each image of the image sequence.
- a model of the pair of glasses 1 1 1 is generated and oriented in an identical manner to the pair of glasses 1 January 1 to create a mask by projection on a layer superimposed on the initial image.
- the appearance of the mask covering the pair of spectacles 1 1 1 is modified in order to erase on the screen the pair of spectacles 1 1 1 worn on the face of the individual.
- a flat map of the environment in the background of the pair of glasses 1 1 1 is created and dynamically updated taking into account the information acquired for each image of the video.
- An inpainting method makes it possible to determine the color of each pixel of the mask of the pair of glasses 1 1 1 as a function of at least one pixel of the image near the pixel of the mask.
- the face is included in the environment of the pair of glasses 1 1 1 but is not detected for the preparation of the map representing the environment. Only the pair of glasses 1 1 1 is detected and followed.
- the presence of the face of the individual 820 is detected but is not followed.
- a model of the face is thus generated and positioned in relation to the position of the pair of glasses followed to the image.
- the model of the face is used in projection for the development of the environmental map.
- the face model can also be directly used by the inpainting method.
- the 820 individual can try a pair of virtual glasses or makeup and see themselves on the screen with. It should be emphasized that in the case of the fitting of a virtual object, only the appearance of the apparent part of the pair of glasses 1 1 1, that is to say not covered by the projection of the virtual object, can be advantageously modified, thus saving computation time.
- FIG. 11 represents a screen 910 displaying a video 915 stored in a computer memory or a real-time video stream from a camera.
- the video 915 shows the head of an individual 920 wearing the pair of glasses 1 1 1 on the face 921 before treatment.
- FIG. 12 represents the screen 910 displaying the video 915 but in which the pair of glasses 11 1 is occulted on each image of the video by a method of generating a final image from an initial image according to FIG. invention.
- the face 921 is detected and tracked on each frame of the video.
- the process adds a layer including an opaque mask covering the pair of glasses 1 1 1.
- the mask is sized to cover most of the shapes and sizes of pairs of spectacles.
- the mask is not linked to the pair of glasses January 1 which is not detected in the present example.
- the method thus generates a layer for each image, on which the mask is oriented and dimensioned in relation to the detected face.
- the generation process applies on the mask a texture from a model of the face previously established without the pair of glasses.
- the method comprises a technique of "relighting" the texture of the mask, making it possible to adjust the colorimetry of the texture to the real light illuminating the face 921.
- Holes can be established on each mask at the level of the eyes of the face 921 to make them visible on each image.
- the holes are not made on the mask when the eyes are optically deformed by the glasses of the pair of spectacles 1 1 1 or when the glasses are tinted.
- the orientation of the synthesized eyes can advantageously be established from the actual orientation of the detected eyes and followed by techniques well known to those skilled in the art.
- a real object to be erased from the image may be a hat, a scarf, hair or any other element partially or completely covering a face.
- the method can also be applied to any other real object that is sought to hide on an image, such as a garment worn by an individual.
- an object to be placed on the face of an individual to replace the pair of glasses worn on the face is makeup, jewelry or even clothing.
- An individual wearing a pair of glasses can thus virtually try a make-up or an evening dress by removing the image of the worn pair of glasses, thus making it possible to simulate the wearing of contact lenses.
- a scan of the body's morphology of the individual may be helpful in obtaining a realistic rendering of the wearing of the garment.
- an individual wearing a pair of glasses is seen on the screen with the same pair of glasses but with a frame having a color, a texture and / or materials different from those of the frame of the pair of glasses actually worn.
- an individual wearing a pair of glasses is seen on the screen with the same pair of glasses but with glasses different in color from the glasses of the pair of glasses actually worn.
- an individual wearing a pair of glasses is seen on the screen with the same pair of glasses but with glasses having a treatment different from that of the glasses of the pair of glasses actually worn.
- the treatment is the addition or removal of one or a combination of treatments well known opticians, such as anti-reflective treatment or thinning glasses.
- an individual wearing a pair of glasses is seen on the screen trying a new pair of virtual glasses where the areas of the glasses of the actual pair of glasses included in the image to the inside the circles of the virtual pair of glasses are preserved, thereby increasing the realism of the virtual pair of glasses.
- the real reflections due to the environment are also preserved in the image.
- the color of the conserved portion of the actual lenses may be changed to obtain a virtual pair of glasses with tinted or untinted lenses, while retaining the actual reflections on the lenses.
- a virtual object is partially superimposed on the real object to be erased from the image and only the visible parts of the corresponding mask of the real object are modified.
- the real object is partially erased from the image or mainly from the image.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Graphics (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Architecture (AREA)
- Processing Or Creating Images (AREA)
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP17742822.4A EP3479344B1 (fr) | 2016-06-30 | 2017-06-29 | Procédé d'occultation d'un objet dans une image ou une vidéo et procédé de réalité augmentée associé |
JP2018569126A JP7112336B2 (ja) | 2016-06-30 | 2017-06-29 | 画像又はビデオ内の物体を隠す方法及び関連する拡張現実方法 |
CN201780053200.5A CN109983501B (zh) | 2016-06-30 | 2017-06-29 | 一种隐藏图像或视频中的对象的方法和相关的增强现实方法 |
KR1020197002130A KR102342982B1 (ko) | 2016-06-30 | 2017-06-29 | 이미지 또는 비디오 내의 객체를 은폐하기 위한 방법 및 관련된 증강 현실 방법 |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
FR1656154 | 2016-06-30 | ||
FR1656154A FR3053509B1 (fr) | 2016-06-30 | 2016-06-30 | Procede d’occultation d’un objet dans une image ou une video et procede de realite augmentee associe |
US15/285,554 | 2016-10-05 | ||
US15/285,554 US9892561B2 (en) | 2016-06-30 | 2016-10-05 | Method of hiding an object in an image or video and associated augmented reality process |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2018002533A1 true WO2018002533A1 (fr) | 2018-01-04 |
Family
ID=59384195
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/FR2017/051744 WO2018002533A1 (fr) | 2016-06-30 | 2017-06-29 | Procédé d'occultation d'un objet dans une image ou une vidéo et procédé de réalité augmentée associé |
Country Status (1)
Country | Link |
---|---|
WO (1) | WO2018002533A1 (fr) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021262187A1 (fr) * | 2020-06-26 | 2021-12-30 | Hewlett-Packard Development Company, L.P. | Ré-éclairage d'image de document |
CN114040129A (zh) * | 2021-11-30 | 2022-02-11 | 北京字节跳动网络技术有限公司 | 视频生成方法、装置、设备及存储介质 |
FR3118821A1 (fr) | 2021-01-13 | 2022-07-15 | Fittingbox | Procédé de détection et de suivi dans un flux vidéo d’un visage d’un individu portant une paire de lunettes |
US12020357B2 (en) | 2020-10-09 | 2024-06-25 | Swimc Llc | Augmentation of digital images with simulated surface coatings |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR1050305A (fr) | 1951-07-31 | 1954-01-06 | Holophane | Appareil d'éclairage direct pour sources lumineuses rectilignes |
FR1551531A (fr) | 1967-12-04 | 1968-12-27 | ||
WO2010042990A1 (fr) * | 2008-10-16 | 2010-04-22 | Seeing Machines Limited | Commercialisation en ligne de produits pour le visage à l'aide d'un suivi du visage en temps réel |
FR2955409A1 (fr) | 2010-01-18 | 2011-07-22 | Fittingbox | Procede d'integration d'un objet virtuel dans des photographies ou video en temps reel |
FR2971873A1 (fr) | 2011-02-22 | 2012-08-24 | Fittingbox | Procede de detection d'un ensemble predefini de points caracteristiques d'un visage |
US20150055085A1 (en) * | 2013-08-22 | 2015-02-26 | Bespoke, Inc. | Method and system to create products |
WO2016020921A1 (fr) * | 2014-08-04 | 2016-02-11 | Pebbles Ltd. | Procédé et système pour la reconstruction de parties occultées d'un visage, dans un environnement de réalité virtuelle |
WO2016050729A1 (fr) * | 2014-09-30 | 2016-04-07 | Thomson Licensing | Retouche faciale à l'aide d'une déformation affine par morceaux et codage épars |
-
2017
- 2017-06-29 WO PCT/FR2017/051744 patent/WO2018002533A1/fr unknown
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR1050305A (fr) | 1951-07-31 | 1954-01-06 | Holophane | Appareil d'éclairage direct pour sources lumineuses rectilignes |
FR1551531A (fr) | 1967-12-04 | 1968-12-27 | ||
WO2010042990A1 (fr) * | 2008-10-16 | 2010-04-22 | Seeing Machines Limited | Commercialisation en ligne de produits pour le visage à l'aide d'un suivi du visage en temps réel |
FR2955409A1 (fr) | 2010-01-18 | 2011-07-22 | Fittingbox | Procede d'integration d'un objet virtuel dans des photographies ou video en temps reel |
FR2971873A1 (fr) | 2011-02-22 | 2012-08-24 | Fittingbox | Procede de detection d'un ensemble predefini de points caracteristiques d'un visage |
EP2678804A1 (fr) | 2011-02-22 | 2014-01-01 | Fittingbox | Procédé de détection d'un ensemble prédéfini de points caractéristiques d'un visage |
US20150055085A1 (en) * | 2013-08-22 | 2015-02-26 | Bespoke, Inc. | Method and system to create products |
WO2016020921A1 (fr) * | 2014-08-04 | 2016-02-11 | Pebbles Ltd. | Procédé et système pour la reconstruction de parties occultées d'un visage, dans un environnement de réalité virtuelle |
WO2016050729A1 (fr) * | 2014-09-30 | 2016-04-07 | Thomson Licensing | Retouche faciale à l'aide d'une déformation affine par morceaux et codage épars |
Non-Patent Citations (4)
Title |
---|
BLANZ; VETTER, FACE RECOGNITION BASED ON FITTING A 3D MORPHABLE MODEL, 2003 |
CANDIDE-3 - AN UPDATED PARAMETERISED FACE, AHLBERG, TECHNICAL REPORT, 2001 |
PAYSAN ET AL., A 3D FACE MODEL FOR POSE AND ILLUMINATION INVARIANT FACE RÉCOGNITION, 2009 |
XIAO-JUN LI ET AL: "Research and Application of Online Product Display Technology Based on Augmented Reality", INFORMATION TECHNOLOGY JOURNAL, vol. 12, no. 6, January 2013 (2013-01-01), PK, pages 1134 - 1142, XP055351955, ISSN: 1812-5638, DOI: 10.3923/itj.2013.1134.1142 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021262187A1 (fr) * | 2020-06-26 | 2021-12-30 | Hewlett-Packard Development Company, L.P. | Ré-éclairage d'image de document |
US12020357B2 (en) | 2020-10-09 | 2024-06-25 | Swimc Llc | Augmentation of digital images with simulated surface coatings |
FR3118821A1 (fr) | 2021-01-13 | 2022-07-15 | Fittingbox | Procédé de détection et de suivi dans un flux vidéo d’un visage d’un individu portant une paire de lunettes |
WO2022153009A1 (fr) | 2021-01-13 | 2022-07-21 | Fittingbox | Procédé de détection et de suivi dans un flux vidéo d'un visage d'un individu portant une paire de lunettes |
CN114040129A (zh) * | 2021-11-30 | 2022-02-11 | 北京字节跳动网络技术有限公司 | 视频生成方法、装置、设备及存储介质 |
CN114040129B (zh) * | 2021-11-30 | 2023-12-05 | 北京字节跳动网络技术有限公司 | 视频生成方法、装置、设备及存储介质 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP3479344B1 (fr) | Procédé d'occultation d'un objet dans une image ou une vidéo et procédé de réalité augmentée associé | |
EP2526510B1 (fr) | Procede de realité augmenté appliquée à l'intégration d'une paire de lunettes dans une image de visage | |
EP3401879B1 (fr) | Procédé de modélisation d'un objet tridimensionnel à partir d'images bidimensionnelles de cet objet prises sous différents angles | |
EP3659109B1 (fr) | Procédé de détermination d'au moins un paramètre associé à un dispositif ophtalmique | |
EP2760329B1 (fr) | Procede de determination de mesures oculaires et optiques | |
FR3067151B1 (fr) | Procede d'essayage virtuel realiste d'une paire de lunettes par un individu | |
ES2742416T3 (es) | Dispositivo y método de imagen corneal | |
CN106575450A (zh) | 通过反照率模型、系统和方法的增强现实内容渲染 | |
EP2455916A1 (fr) | Interface homme-machine basée sur le suivi non-rigide | |
CN112509040A (zh) | 对提供镜面反射和反射修改的表面进行基于图像的检测 | |
JP2017524163A (ja) | 選択された眼鏡フレームの画像データに基づいた使用者データの決定 | |
WO2018002533A1 (fr) | Procédé d'occultation d'un objet dans une image ou une vidéo et procédé de réalité augmentée associé | |
FR3066304A1 (fr) | Procede de compositon d'une image d'un utilisateur immerge dans une scene virtuelle, dispositif, equipement terminal, systeme de realite virtuelle et programme d'ordinateur associes | |
CN110446968A (zh) | 用于确定定心参数的计算机实施的方法 | |
US20230055013A1 (en) | Accessory Detection and Determination for Avatar Enrollment | |
CA3204647A1 (fr) | Procede de detection et de suivi dans un flux video d'un visage d'un individu portant une paire de lunettes | |
WO2020064763A1 (fr) | Détermination automatique des paramètres nécessaires à la réalisation de lunettes | |
WO2008155489A2 (fr) | Procede de synthese d'une image d'iris de-refractee |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17742822 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2018569126 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 20197002130 Country of ref document: KR Kind code of ref document: A |
|
ENP | Entry into the national phase |
Ref document number: 2017742822 Country of ref document: EP Effective date: 20190130 |