EP2132709A1 - Procede et dispositif de determination de la pose d'un objet tridimensionnel dans une image et procede et dispositif de creation d'au moins une image cle pour le suivi d'objets - Google Patents
Procede et dispositif de determination de la pose d'un objet tridimensionnel dans une image et procede et dispositif de creation d'au moins une image cle pour le suivi d'objetsInfo
- Publication number
- EP2132709A1 EP2132709A1 EP08775590A EP08775590A EP2132709A1 EP 2132709 A1 EP2132709 A1 EP 2132709A1 EP 08775590 A EP08775590 A EP 08775590A EP 08775590 A EP08775590 A EP 08775590A EP 2132709 A1 EP2132709 A1 EP 2132709A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- dimensional
- image
- determining
- pose
- generic model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
- G06T7/75—Determining position or orientation of objects or cameras using feature-based methods involving models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
-
- 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
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
Definitions
- the present invention relates to the combination of real and virtual images in real time, also called augmented reality, and more particularly a method and a device for determining the laying of a three-dimensional object in an image and a method and a device for creating at least one keyframe corresponding to a three-dimensional object.
- Augmented reality is intended to insert one or more virtual objects in the images of a video stream.
- the position and orientation of these virtual objects can be determined by data external to the scene represented by the images, for example coordinates directly derived from a game scenario, or by related data. to certain elements of this scene, for example coordinates of a particular point of the scene such as the hand of a player.
- element tracking algorithms also known as target tracking algorithms
- use a marker that can be visual or use other means such as radio frequency or infrared based means.
- some algorithms use pattern recognition to track a particular element in an image stream.
- the Federal Polytechnic School of Lausanne a developed a visual tracking algorithm that does not use a marker and whose originality lies in the pairing of particular points between the current image of a video stream with a keyframe, called a keyframe, given by the user at system boot and an updated keyframe during visual tracking.
- the objective of this visual tracking algorithm is to find, in a real scene, the pose, that is to say the position and the orientation, of an object whose three-dimensional mesh is available, or to find the extrinsic parameters of position and orientation, relative to this object, of a camera filming this object, immobile, thanks to image analysis.
- a keyframe is composed of two elements: a captured image of the video stream and a pose (orientation and position) of a three-dimensional model appearing in this image.
- Keyframes should be "offline", off-line, or "on-line” keyframes, or on one.
- Offline keyframes are images extracted from the video stream in which the object to be tracked has been placed manually through the use of a pointing device such as a mouse or using an adjustment tool such as a Pocket Dial marketed by the company Doepfer.
- the offline keyframes preferably characterize the pose of the same object in several images.
- “Online” keyframes are dynamically stored during the execution of the tracking program. They are calculated when the error, that is the distance between the points of interest pairings, is small. Online keyframes replace the offline keyframes used to initialize the application. Their use aims to reduce the shift, also called drift, which can become important when one moves too far from the initial relative position between the camera and the object. Learning new online keyframes also results in more robust application to external light variations and camera color variations. However, they have the disadvantage of introducing a "vibration" effect on the pose of the object over time. When learning a new online keyframe, this one just replace the previous keyframe, offline or online. It is used as a common keyframe.
- Each keyframe includes an image in which the object is present and a pose to characterize the location of that object as well as a number of points of interest that characterize the object in the object. picture.
- Points of interest are, for example, constructed from a Harris point detector and represent locations with high values of directional gradients in the image.
- the manual preparation phase thus consists in finding a first estimate of the pose of the object in an image extracted from the video stream, which amounts to formalizing the initial affine transformation T p ⁇ c which corresponds to the matrix of passage between the reference mark attached to the object to the mark associated with the camera.
- the transformation T p ⁇ 0 can then be expressed in the form of the following matrix, cos cos cos + sin ps sin sin sin? cos-cos-cos ⁇ sin a cos ⁇ sin a rp-cos> cos? cos? cos sin sin # ⁇ p sin cosin # cos - cos z cos sin ⁇ - o, os ⁇ sm. ⁇ cos ⁇ -sm. ⁇ sm. ⁇ cos # cos ⁇ ⁇ ⁇
- this model makes it possible to establish the link between the coordinates of the points of the three-dimensional model of the object expressed in the reference of the object and the coordinates of these points in the reference of the camera.
- the offline keyframes are processed to position points of interest according to the parameters chosen at the launch of the application. These parameters are empirically specified for each type of application use and allow the matching detection core to be modulated and to obtain a better quality in estimating the pose of the object according to the characteristics of the application. the real environment. Then, when the real object in the current image is in a pose that is close to the pose of the same object in one of the offline keyframes, the number of matches becomes important. It is then possible to find the affine transformation allowing to fix the virtual three-dimensional model of the object on the real object.
- the algorithm goes into steady state.
- the displacements of the object are followed by one frame on the other and any drifts are compensated thanks to the information contained in the offline key image retained during the initialization and in the online key image calculated during the execution of the application.
- the tracking application combines two types of algorithms: a detection of points of interest, for example a modified version of Harris point detection, and a reprojection technique of the points of interest positioned on the three-dimensional model towards the flat image. This reprojection makes it possible to predict the result of a spatial transformation from one frame to the other.
- a detection of points of interest for example a modified version of Harris point detection
- a reprojection technique of the points of interest positioned on the three-dimensional model towards the flat image. This reprojection makes it possible to predict the result of a spatial transformation from one frame to the other.
- a point p of the image is the projection of a point P of the real scene with p ⁇ P, - P E • T p ⁇ c - P
- Pi is the matrix of intrinsic parameters of the camera, ie its focal length, the center of the image and the offset
- PE is the matrix of the extrinsic parameters of the camera, ie the position of the camera in the real space
- T p ⁇ c the affine matrix of passage between the reference associated with the object followed towards the reference of the camera. Only the relative position of the object relative to the relative position of the camera is considered here, which amounts to placing the reference of the real scene at the optical center of the camera.
- the pose of an object is estimated according to the correspondences between the points of interest of the current image resulting from the video stream, the points of interest of the current key image and the points of interest of the previous image. of the video stream. These operations are called the matching phase. From the most significant correlations, the software calculates the pose of the object corresponding best to the observations.
- FIGS 1 and 2 illustrate this tracking application.
- the proposed solutions are often derived from research and do not take into account the constraints of implementation of commercial systems.
- the problems related to robustness, the ability to quickly launch the application without requiring a manual phase of creation of one or more keyframes off-line necessary for the initialization of the tracking system, the detection of "off-hook" errors (when the object to be tracked is “lost") and automatic and real-time reset after such errors are often left out.
- the invention solves at least one of the problems discussed above.
- the invention thus relates to a method for determining the laying of a three-dimensional object in an image, characterized in that it comprises the following steps:
- the method according to the invention thus makes it possible to automatically determine the pose of a three-dimensional object in an image, in particular to create key initialization images of an augmented reality application using automatic tracking, in real time, of three-dimensional objects in a video stream. This determination is based on the acquisition of a model of the object and the projection of it according to at least a two-dimensional representation, then on a positioning of a representation on the object in the image in order to determine the pose.
- the method comprises a preliminary step of constructing a three-dimensional generic model of the object from the three-dimensional object.
- the three-dimensional generic model is a mesh of the object.
- the method comprises a preliminary step of three-dimensional localization of the object in the image. This feature makes it easier to position a two-dimensional representation of the object in the image.
- the method comprises a step of determining the characteristic points of the object of the image.
- This feature thus facilitates the positioning of a two-dimensional representation of the object in the image and the determination of the three-dimensional pose of an object in an image when a two-dimensional representation is positioned.
- the method comprises a preliminary step of determining characteristic points of the three-dimensional generic model of the object. According to this feature, the positioning of a two-dimensional representation is facilitated as well as the determination of the three-dimensional pose of an object in an image when a two-dimensional representation is positioned.
- the step of determining the three-dimensional pose of the object in the image is furthermore a function of the distance between the determined characteristic points of the three-dimensional generic model of the object and the determined characteristic points of the object. object in the image.
- the invention also relates to a method for creating at least one key image comprising an image representing at least one three-dimensional object in a three-dimensional environment, this method being characterized in that it comprises the following steps:
- the method according to the invention thus makes it possible to automate the creation of key images, in particular with a view to initializing or resetting an application of augmented reality using automatic, real-time tracking of three-dimensional objects in a video stream.
- the invention also relates to a device for determining the pose of a three-dimensional object in an image, characterized in that it comprises the following means:
- projection means of the three-dimensional generic model according to at least one two-dimensional representation and means of association with each two-dimensional representation of a pose information of the three-dimensional object
- the invention proposes a device for creating at least one key image comprising an image representing at least one three-dimensional object in a three-dimensional environment, this device being characterized in that it comprises the following means:
- the present invention also aims a storage means, removable or not, partially or completely readable by a computer or a microprocessor comprising code instructions of a computer program for executing each of the steps of the methods as set forth above.
- the present invention provides a computer program comprising instructions adapted to the implementation of each of the steps of the methods as described above.
- FIG. 1 schematically represents the essential principles of the object tracking application developed by the Swiss Federal Institute of Technology in Lausanne;
- FIG. 2 illustrates certain steps of the method for determining the placing of an object in an image of a video stream from key images and the previous image of the video stream;
- FIG. 3 represents the global diagram of the creation of one or more key images of a three-dimensional object and of any geometry, in an environment implementing the invention;
- FIG. 4 shows an example of an apparatus making it possible to implement the invention at least partially;
- FIG. 5 illustrates a generic algorithm based on the image analysis according to the invention;
- FIG. 6 illustrates an image analysis algorithm for creating key images of a face according to the invention.
- FIG. 7 illustrates the creation of two-dimensional models obtained from the generic three-dimensional mesh of a face.
- the method according to the invention particularly relates to the creation, in particular automatically, of at least one key image of at least one three-dimensional object in an environment with a view to automating the initialization and reset phases after a stall. the object tracking application on images from a video stream.
- a key image is sufficient for the automation of the initialization and reset phases, especially when the pose of the object in an image is found in real time and very precisely by means of an analysis. image.
- a multitude of keyframes may also allow the application to initialize for any type of relative poses between the object to be tracked and the camera.
- FIG. 3 illustrates the global scheme for creating one or more keyframes, also called initialization keyframes of an object in an environment implementing the invention, for an object tracking application.
- the creation of at least one key image of an object in an environment and the execution of a tracking application (300) using these keyframes comprise three interconnected phases: a creation phase one or more initialization keyframes (I), a tracking initialization phase that use the initialization keyframe (s) previously create (II) and an object tracking phase (III) corresponding to the steady state of the application and that can be used to create new initialization keyframes.
- the phase of creating a first keyframe (I) consists mainly of acquiring an image representing the three-dimensional object in an initial position. This acquisition is performed, in particular, from a shooting means such as a camera or a camera.
- a first key image is created (step 320) comprising on the one hand the first image acquired and the relative pose of the object in the environment according to the point view of the image.
- an image analysis module is thus introduced prior to the creation of the key image (step 310) and makes it possible to find the pose of the object in the image without user intervention. .
- a prior knowledge of the type of object to be found in the image and a knowledge of some characteristics thereof can indeed estimate the pose of the object in space real. This approach is particularly interesting when it comes for example to find the pose of a face in an image. Indeed, it is possible to use facial features such as the eyes or the mouth, to determine the pose of the object.
- phase I the steps of this phase I can be reiterated for the creation of a plurality of key images, without requiring the intervention of the user.
- the tracking application is initialized by searching for a keyframe representing the object in the stream video containing the object to follow (step 320).
- the tracking application can find the object (phase III) in the successive images of the stream video according to a tracking mechanism (step 325).
- a tracking mechanism According to this mechanism, the displacements of the object (displacement of the object in the scene or displacement induced by the movement of the camera in the scene) are followed by a frame on the other and any drifts are compensated thanks to the information contained in the initialization keyframe retained during initialization and, optionally, in the initialization keyframe computed during the execution of the application.
- These keyframes themselves can later be used as an initialization keyframe to re-initialize the application automatically.
- the phase of reset is similar to the initialization phase described above (step 320).
- this scheme of creating one or more keyframes can be repeated to allow the creation of new keyframes corresponding to other objects also present in the image. Once the creation of at least one keyframe for each object is complete, it is possible to track multiple objects in the video stream.
- Figure 4 shows schematically an apparatus adapted to implement the invention.
- the device 400 is for example a microcomputer, a workstation or a game console.
- the device 400 preferably comprises a communication bus 402 to which are connected:
- a central processing unit or microprocessor 404 CPU, Central Processing Unit
- ROM 406 Read OnIy Memory
- Programs such as "Prog"
- RAM 088 Random Access Memory
- a video acquisition card 410 connected to a camera 412;
- a graphics card 416 connected to a screen or to a projector 418.
- the apparatus 400 may also have the following elements:
- a hard disk 420 which may comprise the aforementioned "Prog" programs and data processed or to be processed according to the invention
- a distributed communication network 428 for example the Internet network
- a data acquisition card 414 connected to a sensor (not shown);
- a memory card reader (not shown) adapted to read or write processed or processed data according to the invention.
- the communication bus allows communication and interoperability between the various elements included in the device 400 or connected to it.
- the representation of the bus is not limiting and, in particular, the central unit is capable of communicating instructions to any element of the apparatus 400 directly or via another element of the apparatus 400.
- the executable code of each program enabling the programmable device to implement the processes according to the invention can be stored, for example, in the hard disk 420 or in the read-only memory 406.
- the executable code of the programs may be received via the communication network 428, via the interface 426, to be stored in the same manner as that described previously.
- the memory cards can be replaced by any information medium such as, for example, a compact disc (CD-ROM or DVD).
- the memory cards can be replaced by information storage means, readable by a computer or by a microprocessor, integrated or not integrated into the device, possibly removable, and adapted to store one or more programs including the execution allows the implementation of the method according to the invention.
- the program or programs may be loaded into one of the storage means of the device 400 before being executed.
- the central unit 404 will control and direct the execution of the instructions or portions of software code of the program or programs according to the invention, instructions which are stored in the hard disk 420 or in the ROM 406 or in the other storage elements mentioned above.
- the program or programs that are stored in a non-volatile memory for example the hard disk 420 or the read only memory 406, are transferred into the random access memory 408 which then contains the executable code of the program or programs according to the invention, as well as registers for storing the variables and parameters necessary for the implementation of the invention.
- the communication apparatus comprising the device according to the invention can also be a programmed apparatus.
- This device then contains the code of the computer program or programs for example frozen in a specific application integrated circuit (ASIC).
- ASIC application integrated circuit
- the image from the video card 416 can be transmitted to the screen or the projector 418 through the communication interface 426 and the distributed communication network 428.
- the camera 412 can be connected to a card video acquisition device 410 ', separate from the camera 400, so that the images from the camera 412 are transmitted to the camera 400 through the distributed communication network 428 and the communication interface 426.
- a tracking application can be initialized from this set and used in a standard way to track an object in a sequence of images from a video stream, for example. example to embed a video sequence on an object of the scene by taking into account the position and orientation of this object, but also to determine the movement of a camera according to the analysis of an object of the scene. In this case the object is part of the decor. Find the pose of this object in the scene is therefore to find the pose of the camera in relation to it. It then becomes possible to add virtual elements in the scene provided that the geometric transformation between the object and the geometric model of the scene is known. What is the case. This approach allows to increase the real scene with animated virtual objects that move according to the geometry of the scene.
- This algorithm comprises in particular two phases, one of which can be carried out offline and the other being carried out online. This second phase executing, in particular, at each phase of initialization or reinitialization of the object tracking algorithm.
- steps 505 to 520 consist first of all to obtain the knowledge concerning the shape of the object to be followed in the image (steps 505).
- This knowledge is, in particular, related to the type of objects to follow in the video stream.
- this knowledge may relate to one or more objects faces to locate in any environment, or one or more trees in a landscape.
- the three-dimensional generic model of the object is constructed from a generic form of the real object, in particular from the real object to be found in the video stream, the latter possibly being the mesh. of the object. On this mesh, it is identified and positioned, in step 515, including manually, the characteristic elements of the object. Regarding the face, these include the nose, eyes and mouth.
- the mesh with its identified characteristic elements is projected according to one or more two-dimensional representations, and to each of these representations, it is associated with a pose information of the three-dimensional object represented (step 520).
- Each two-dimensional representation thus corresponds to a pose that the three-dimensional object can have.
- the mesh is sampled according to a plurality of possible positions, orientations and scales.
- several models corresponding to various random values or not (depending on the use case) in the parameter space are constructed. These parameters are defined in particular in the three-dimensional space.
- This space includes the following orientation parameters: yaw (yaw) in terminology corresponding to the rotation around the axis z -> ⁇ , the pitch ("pitch” in English terminology) corresponding to the rotation around the axis x -> ⁇ and the roll (“millet”) in English terminology) corresponding to the rotation around the axis y -> ⁇ .
- yaw yaw
- the pitch in English terminology
- the roll corresponding to the rotation around the axis y -> ⁇
- the set of two-dimensional representations of the previously generated three-dimensional object is made available (step 525).
- an image is extracted (step 530).
- the object in the image is possibly located approximately in two dimensions or three dimensions (the size of the object in the image can give us depth information), (step 535).
- the technique of discretized Haar wavelets makes it possible to search the image for a model similar to that previously learned on hundreds of objects of the same type with small differences.
- it is for example identified a frame encompassing the object or objects to look for in the image, and possibly parts thereof.
- step 540 of searching for characteristic elements of the object in the image is followed by step 540 of searching for characteristic elements of the object in the image.
- These characteristic elements can be points, segments and curves that belong to the object. From these elements important information about the position and orientation of the object can be deduced. To do this the use of image analysis methods is relevant. For example, the following operations can be performed: gradient analyzes, the determination of colorimetric thresholds in different color spaces, the application of filters, for example the LoG filter ("Laplacian of Gaussian” in English terminology) or the Sobel filter, energy minimizations, especially the extraction of contours ("snake” in English terminology) taking into account, for example, the color of the object to be found in the image to find its outline in two dimensions.
- gradient analyzes the determination of colorimetric thresholds in different color spaces
- filters for example the LoG filter ("Laplacian of Gaussian" in English terminology) or the Sobel filter
- energy minimizations especially the extraction of contours ("snake” in English terminology) taking into account, for example, the color of the object to be found in the image to find its outline in two dimensions.
- the positioning corresponds in particular to search for correspondence between the two-dimensional representation and the object in the image
- the pose is determined at least from the pose information associated with the selected two-dimensional representation.
- the pose is determined from the distance between the characteristic elements found.
- the pose of the three-dimensional object is determined, including in particular the orientation and the position of the object. This information is used to create an initialization keyframe for the real-time object tracking application in a video stream in step 320 of Figure 3.
- the application may consist of finding the pose of a three-dimensional object, for example a face, in a real-time video stream in order to enrich the object tracked.
- a three-dimensional object for example a face
- Such an application works for any type of "face" type object present in the video stream.
- the user can, for example see, through a control screen, his face enriched by various three-dimensional objects of synthesis, especially on his face real, it is added a hat or glasses.
- the user can look like known virtual characters or a character of his choice that he previously modeled.
- the method of the invention makes it possible to carry out the placement of the mesh automatically at the launch of the application.
- the mesh corresponding to a generic face is previously modeled using data of general proportions of a human face. These proportions being very close to one person to another, the algorithm is made more robust to the different users likely to interact with the application.
- the initialization phase is made automatic, in particular by the use of an image analysis solution to find certain characteristic points in the image.
- the characteristic points may be the eyes, the mouth, the nose, the eyebrows and the chin.
- This information here described in a non-exhaustive manner and identified according to the type of application to be implemented, makes it possible to position the mesh corresponding to the face automatically, realistically and accurately. Indeed, any offset in the estimate of the initial position of the mesh on the face would then be very harmful during the execution of the tracking application.
- FIG. 6 illustrates an image analysis algorithm for creating key images of a face according to the invention.
- the search by image analysis firstly makes it possible to accurately find the position of a face in the images of a video stream (step 600).
- this image analysis uses, for example, the technique of discretized Haar wavelets to search the image for a model similar to that previously learned on hundreds of different faces.
- it is identified a frame encompassing the face determined in the image and possibly narrower frames identifying certain characteristic areas of the face (for example around the eyes or mouth) which will then allow a search more precise elements of the face. This first approximation of the position of certain elements of the face may be insufficient to find these characteristic elements accurately.
- the next step (step 605) is to determine more precisely in these regions points, segments and characteristic curves that belong to the face and gives us important information on the position and orientation of the face. These include, for example, eyes, eyebrows, mouth and the axis of the nose. These elements are found by means of image analysis.
- the following operations are performed: gradient analysis, simple pattern recognition (ovoid around the eyes), the determination of colorimetric thresholds (such as those that characterize the hue of a mouth), l application of filters, for example the LoG filter ("Laplacian of Gaussian", to accentuate the contours present in the face) or the Sobel filter (to find characteristic points), energy minimizations, in particular the extraction of contours ("Snake”) taking into account, for example, the hue of the skin.
- filters for example the LoG filter ("Laplacian of Gaussian", to accentuate the contours present in the face) or the Sobel filter (to find characteristic points)
- energy minimizations in particular the extraction of contours ("Snake") taking into account, for example, the hue of the skin.
- the mesh is sampled according to a plurality of possible positions, orientations and scales.
- several models corresponding to various random values or not (depending on the use case) in the parameter space are constructed. These parameters are defined in particular in the three-dimensional space. This space includes the following orientation parameters: yaw, pitch, and roll. These settings may vary very slightly. Indeed, it is considered that the user is more or less correctly positioned in front of the camera.
- This learning step thus makes it possible to create a series of simplified and projected two-dimensional models obtained from the generic three-dimensional mesh as illustrated in FIG. 7, for which three-dimensional laying parameters are associated so as to create keyframes.
- the algorithm continues with the adjustment of the generic mesh (step 610) and the automatic search for the pose of the three-dimensional object (step 615). To do this, from the set of projections corresponding to the generic mesh and relevant information of the face, the mesh is adjusted and the pose of the three-dimensional object in the recovered image.
- the projected and simplified meshes are thus compared with the relevant information of the face, namely the points, the segments and the curves, by means of distance functions.
- a correlation operation will make it possible to estimate the pose and the scaling of the user's face in the initial image extracted from the video stream. Indeed, all the relevant parameters to find the matrix of passage between the three-dimensional generic mesh and the mesh used for monitoring are known.
- the pose of the face in the image extracted from the video stream is known, it is created a first initialization keyframe.
- This can be used directly to allow automatic initialization of the face tracking application. Indeed, each time the user is close to the pose contained in this keyframe, the initialization takes place. However, it remains possible to create several initialization keyframes to allow more robust initialization in various relative poses of the face relative to the camera.
- the user may for example be forced to face the camera during the creation of the image key. In this way, the variation of the degrees of freedom is reduced during the automatic search for the pose of the object. To do this, it is also possible to add targets on the screen that make it possible to force the user to position himself correctly in front of the camera.
- the transformation matrix between the initial position of the generic mesh and the modified position can be expressed by the expression: SRT where S is the scaling matrix, R is the rotation matrix and 7Ia translation matrix.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Graphics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Processing Or Creating Images (AREA)
- Image Analysis (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
FR0753482A FR2913128B1 (fr) | 2007-02-23 | 2007-02-23 | Procede et dispositif de determination de la pose d'un objet tridimensionnel dans une image et procede et dispositif de creation d'au moins une image cle |
PCT/FR2008/000236 WO2008125754A1 (fr) | 2007-02-23 | 2008-02-22 | Procede et dispositif de determination de la pose d'un objet tridimensionnel dans une image et procede et dispositif de creation d'au moins une image cle pour le suivi d'objets |
Publications (1)
Publication Number | Publication Date |
---|---|
EP2132709A1 true EP2132709A1 (fr) | 2009-12-16 |
Family
ID=38606701
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP08775590A Ceased EP2132709A1 (fr) | 2007-02-23 | 2008-02-22 | Procede et dispositif de determination de la pose d'un objet tridimensionnel dans une image et procede et dispositif de creation d'au moins une image cle pour le suivi d'objets |
Country Status (6)
Country | Link |
---|---|
US (1) | US8675972B2 (fr) |
EP (1) | EP2132709A1 (fr) |
JP (1) | JP2010519629A (fr) |
KR (1) | KR20090114471A (fr) |
FR (1) | FR2913128B1 (fr) |
WO (1) | WO2008125754A1 (fr) |
Families Citing this family (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102009049073A1 (de) * | 2009-10-12 | 2011-04-21 | Metaio Gmbh | Verfahren zur Darstellung von virtueller Information in einer Ansicht einer realen Umgebung |
KR101487944B1 (ko) | 2010-02-24 | 2015-01-30 | 아이피플렉 홀딩스 코포레이션 | 시각 장애인들을 지원하는 증강 현실 파노라마 |
US8509522B2 (en) * | 2010-10-15 | 2013-08-13 | Autodesk, Inc. | Camera translation using rotation from device |
US20130010068A1 (en) * | 2011-04-12 | 2013-01-10 | Radiation Monitoring Devices, Inc. | Augmented reality system |
JP5855416B2 (ja) * | 2011-10-25 | 2016-02-09 | Kddi株式会社 | 3次元座標取得装置、カメラ姿勢推定装置、プログラム |
JP5995304B2 (ja) * | 2011-12-12 | 2016-09-21 | 株式会社バンダイナムコエンターテインメント | プログラム、情報記憶媒体、端末及びサーバ |
JP5735455B2 (ja) * | 2012-05-22 | 2015-06-17 | 日本電信電話株式会社 | 移動物体ポーズ算出装置、移動物体ポーズ算出方法、及びプログラム |
US9437011B2 (en) | 2012-06-11 | 2016-09-06 | Samsung Electronics Co., Ltd. | Method and apparatus for estimating a pose of a head for a person |
US10013623B2 (en) * | 2012-06-29 | 2018-07-03 | Blackberry Limited | System and method for determining the position of an object displaying media content |
EP2704055A1 (fr) * | 2012-08-31 | 2014-03-05 | Layar B.V. | Détermination d'espace pour afficher un contenu dans la réalité améliorée |
US9928874B2 (en) | 2014-02-05 | 2018-03-27 | Snap Inc. | Method for real-time video processing involving changing features of an object in the video |
US9892543B2 (en) | 2016-02-26 | 2018-02-13 | Carnegie Mellon University | Systems and methods for estimating pose of textureless objects |
JP6872742B2 (ja) * | 2016-06-30 | 2021-05-19 | 学校法人明治大学 | 顔画像処理システム、顔画像処理方法及び顔画像処理プログラム |
CN108345821B (zh) * | 2017-01-24 | 2022-03-08 | 成都理想境界科技有限公司 | 面部追踪方法及设备 |
EP3693924B1 (fr) * | 2017-10-03 | 2023-08-02 | Fujitsu Limited | Système de reconnaissance de posture, programme de correction d'image et procédé de correction d'image |
WO2019161229A1 (fr) * | 2018-02-15 | 2019-08-22 | DMAI, Inc. | Système et procédé de reconstruction d'espace 3d non occupé |
US11653072B2 (en) * | 2018-09-12 | 2023-05-16 | Zuma Beach Ip Pty Ltd | Method and system for generating interactive media content |
CN110148092B (zh) * | 2019-04-16 | 2022-12-13 | 无锡海鸿信息技术有限公司 | 基于机器视觉的青少年坐姿及情绪状态的分析方法 |
CN110111428B (zh) * | 2019-05-28 | 2023-06-20 | 艾瑞迈迪科技石家庄有限公司 | 一种应用于增强现实的虚拟目标标定方法及装置 |
CN112233018B (zh) * | 2020-09-22 | 2023-01-06 | 天津大学 | 基于三维形变模型的参考图引导的人脸超分辨率方法 |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6207474B1 (en) * | 1998-03-09 | 2001-03-27 | Micron Technology, Inc. | Method of forming a stack of packaged memory die and resulting apparatus |
US6671391B1 (en) * | 2000-05-26 | 2003-12-30 | Microsoft Corp. | Pose-adaptive face detection system and process |
JP4573085B2 (ja) * | 2001-08-10 | 2010-11-04 | 日本電気株式会社 | 位置姿勢認識装置とその位置姿勢認識方法、及び位置姿勢認識プログラム |
JP2003308533A (ja) * | 2002-04-17 | 2003-10-31 | Mitsubishi Electric Corp | ドライバの顔画像処理装置 |
WO2005086089A1 (fr) * | 2004-03-03 | 2005-09-15 | Nec Corporation | Système d'estimation/corrélation de posture d'objet, méthode d'estimation/corrélation de posture d'objet et programme pour ceux-ci |
JP4677273B2 (ja) * | 2004-06-29 | 2011-04-27 | キヤノン株式会社 | 情報処理方法および情報処理装置 |
US7536030B2 (en) * | 2005-11-30 | 2009-05-19 | Microsoft Corporation | Real-time Bayesian 3D pose tracking |
-
2007
- 2007-02-23 FR FR0753482A patent/FR2913128B1/fr active Active
-
2008
- 2008-02-22 KR KR1020097019828A patent/KR20090114471A/ko not_active Application Discontinuation
- 2008-02-22 EP EP08775590A patent/EP2132709A1/fr not_active Ceased
- 2008-02-22 US US12/528,418 patent/US8675972B2/en active Active
- 2008-02-22 WO PCT/FR2008/000236 patent/WO2008125754A1/fr active Application Filing
- 2008-02-22 JP JP2009550305A patent/JP2010519629A/ja active Pending
Non-Patent Citations (1)
Title |
---|
See references of WO2008125754A1 * |
Also Published As
Publication number | Publication date |
---|---|
FR2913128B1 (fr) | 2009-08-28 |
FR2913128A1 (fr) | 2008-08-29 |
KR20090114471A (ko) | 2009-11-03 |
US8675972B2 (en) | 2014-03-18 |
US20100316281A1 (en) | 2010-12-16 |
JP2010519629A (ja) | 2010-06-03 |
WO2008125754A1 (fr) | 2008-10-23 |
WO2008125754A8 (fr) | 2009-09-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP2132709A1 (fr) | Procede et dispositif de determination de la pose d'un objet tridimensionnel dans une image et procede et dispositif de creation d'au moins une image cle pour le suivi d'objets | |
EP2132710B1 (fr) | Procede et dispositifs de realite augmentee utilisant un suivi automatique, en temps reel, d'objets geometriques planaires textures, sans marqueur, dans un flux video | |
EP2111605B1 (fr) | Procédé et dispositif de création d'au moins deux images clés correspondant à un objet tridimensionnel | |
EP2715662B1 (fr) | Procede de localisation d'une camera et de reconstruction 3d dans un environnement partiellement connu | |
EP2455916B1 (fr) | Interface homme-machine basée sur le suivi non-rigide | |
EP2104925B1 (fr) | Procédé et dispositifs pour insérer en temps réel des objets virtuels dans un flux d'images à partir de données issues de la scène réelle représentée par ces images | |
US10580148B2 (en) | Graphical coordinate system transform for video frames | |
FR2933218A1 (fr) | Procede et dispositif permettant de detecter en temps reel des interactions entre un utilisateur et une scene de realite augmentee | |
FR2980681A3 (fr) | Procede de determination des mesures oculaires et optiques d'interet pour la fabrication et le montage de verre de lunettes correcteurs, a l'aide d'une camera en contexte non controle | |
Porzi et al. | Learning contours for automatic annotations of mountains pictures on a smartphone | |
WO2015177461A1 (fr) | Procédé de détermination d'au moins un paramètre comportemental | |
EP2441048A1 (fr) | Procédés et dispositifs d'identification d'objets réels, de suivi de la représentation de ces objets et de réalité augmentée, dans une séquence d'images, en mode client-serveur | |
EP2441046A2 (fr) | Procédé et dispositif de calibration d'un capteur d'images utilisant un système temps réel de suivi d'objets dans une séquence d'images | |
EP2441047A1 (fr) | Procédé et dispositif de suivi d'objets en temps réel dans une séquence d'images en présence de flou optique | |
FR3054062A1 (fr) | Systeme et procede de capture embarquee et de reproduction 3d/360° du mouvement d'un operateur dans son environnement | |
WO2023099302A1 (fr) | Procédé de gestion de la représentation de points d'intérêts | |
FR3112639A1 (fr) | Procédé, dispositif et programme d’ordinateur d’aide au positionnement pour système de réalité étendue | |
FR3052287B1 (fr) | Construction d'une image tridimensionnelle |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
17P | Request for examination filed |
Effective date: 20090911 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MT NL NO PL PT RO SE SI SK TR |
|
17Q | First examination report despatched |
Effective date: 20100408 |
|
DAX | Request for extension of the european patent (deleted) | ||
APBK | Appeal reference recorded |
Free format text: ORIGINAL CODE: EPIDOSNREFNE |
|
APBN | Date of receipt of notice of appeal recorded |
Free format text: ORIGINAL CODE: EPIDOSNNOA2E |
|
APBR | Date of receipt of statement of grounds of appeal recorded |
Free format text: ORIGINAL CODE: EPIDOSNNOA3E |
|
APAF | Appeal reference modified |
Free format text: ORIGINAL CODE: EPIDOSCREFNE |
|
RAP1 | Party data changed (applicant data changed or rights of an application transferred) |
Owner name: QUALCOMM CONNECTED EXPERIENCES, INC. |
|
RAP1 | Party data changed (applicant data changed or rights of an application transferred) |
Owner name: QUALCOMM INCORPORATED |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R003 |
|
APBT | Appeal procedure closed |
Free format text: ORIGINAL CODE: EPIDOSNNOA9E |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE APPLICATION HAS BEEN REFUSED |
|
18R | Application refused |
Effective date: 20200122 |