Method and apparatus for adjusting an internal space camera for outer-in tracking of

Abstract

The invention relates to a method for providing camera position data which indicates a camera position of an adjustable interior camera, comprising the following steps: acquiring a camera image of the interior camera; providing a reference camera image for a predetermined reference camera pose; performing an image comparison between the reference camera image and the captured camera image in order to identify feature point pairs having feature points associated with the same element and to ascertain 2D image coordinates in the reference camera image and 2D image coordinates in the captured camera image; according to the 2D image coordinate in the reference camera image and the 2D image coordinate in the collected camera image, determining the camera pose change; and acting on the reference camera pose with the camera pose change to obtain camera pose data.

Description

Technical Field

[0001] The invention relates to the use of data glasses in vehicles, and in particular for displaying contact simulations of display objects. The invention also relates to means for providing the glasses pose by an external system for sensor fusion in the data glasses. Background Art

[0002] Data glasses, also called head-mounted displays, are known which can display images on one or two display surfaces in the field of vision of the wearer of the data glasses by means of a display device. These display surfaces are generally reflective surfaces which direct the images into the eyes of the wearer of the data glasses. The viewing aperture of the data glasses is transparent so that the real environment can be perceived through the data glasses in the usual way. The display surface is located in the viewing aperture so that information to be displayed, such as texts, symbols, graphics, video displays, etc., can be displayed superimposed on the perception of the environment.

[0003] The information is generally displayed to the wearer of the data glasses in a contact-analog manner, i.e., in such a way that the information is superimposed as object information on a specific associated object in the real environment or is positioned in accordance with the specific associated object, or in such a way that the object information to be displayed is displayed in a specific direction of the data glasses or their wearer. The contact-analog object information can also be displayed in such a way that it appears perspectively correct with respect to the objects in the real environment, i.e., the illusion is created that the objects in the real environment have actually been supplemented by additional features of the visual object information.

[0004] In order to display the object information on the display surface of the data glasses in a corresponding contact-simulated manner, it is necessary to know the position of the object in the environment and the position of the user's eyes and the user's line of sight. When wearing data glasses, the user's line of sight is fixedly assigned to the glasses' posture, i.e., the 3D position and 3D orientation of the data glasses with respect to a reference coordinate system, such as a vehicle coordinate system (vehicle-fixed) or an environment coordinate system (world-fixed).

[0005] In particular, when data glasses are used in vehicles, a posture recognition unit can be provided in the data glasses to determine glasses posture data regarding the glasses posture of the data glasses. The posture recognition unit generally has a glasses motion sensing mechanism and a computing device, such as a computing device in the form of a microprocessor. With the aid of glasses motion information detected at a high frequency, in particular glasses motion information in the form of 3D acceleration information and / or 3D angular velocity information, the current glasses posture of the data glasses in the interior space of the vehicle can be tracked by time integration of the relative glasses motion information to update the glasses posture data. Here, the current glasses posture is first updated and re-determined with respect to the world environment. Although a single time integration is sufficient for re-determination of the absolute orientation of the data glasses, a secondary time integration of the 3D translation acceleration information is required for re-determination of the absolute position of the data glasses.

[0006] When determining the absolute eyeglass position data by taking into account the integration of the eyeglass movement information, integration errors generally occur, which, if not corrected, lead to an increasing deviation of the absolute eyeglass position data provided in the data eyeglasses from the actual current eyeglass position. In particular, due to the second integration, the integration error in determining the absolute position of the data eyeglasses is significant.

[0007] In order to find out the current glasses pose data with respect to the vehicle coordinate system, the movement information of the data glasses acquired in the vehicle can be transmitted to the data glasses, so that the absolute glasses pose data with respect to the vehicle (with respect to the vehicle coordinate system) can be determined.

[0008] The current eyewear pose can be determined by an external pose recognition unit (outside-in tracking method: Outside-In-Tracking-Verfahren), in which a camera in the vehicle interior detects the head of the wearer of the data glasses and, by evaluating the camera image with the aid of a visual tracking method, either: the pose of the head is determined and absolute eyewear pose data are derived therefrom; or the absolute eyewear pose data are determined directly. In these so-called outside-in tracking systems, absolute eyewear pose data are determined in relation to the vehicle, i.e., in relation to a vehicle coordinate system fixed to the vehicle.

[0009] The absolute glasses pose data can be transmitted to the data glasses cyclically (with time stamp), and the re-determined absolute glasses pose data obtained by the integration method in the data glasses can be corrected there by sensor fusion to compensate for the integration error.

[0010] For the outside-in tracking of the data glasses, it can be provided that a corresponding interior camera that can be used to determine the absolute fixed glasses position of the vehicle is arranged in the interior rearview mirror unit. However, depending on the driver's body shape and seat position, the interior rearview mirror unit is generally adjusted at the beginning of the journey so that the driver can see the desired section of the rear vehicle environment through the mirror from his perspective. Since the interior camera is arranged in the interior rearview mirror unit, each time the interior rearview mirror unit is readjusted, the interior camera located therein will obtain a new orientation, i.e. a new camera position. However, in order to determine the absolute glasses position, it is necessary to accurately know the orientation and even the position of the interior camera with respect to the vehicle coordinate system. Summary of the invention

[0011] It is therefore an object of the present invention to provide a method for determining precise camera pose data of an interior camera with respect to a vehicle coordinate system.

[0012] This object is achieved by a method for determining absolute camera pose data of an interior camera for ascertaining an absolute eyewear pose in a vehicle according to claim 1 and by a device and a vehicle system according to the independent claims.

[0013] Further refinements are given in the dependent claims.

[0014] According to a first aspect, a method for providing camera pose data is provided, which camera pose data describe the camera pose of an adjustable interior camera, the method comprising the following steps:

[0015] - Acquisition of camera images of the interior space camera;

[0016] - providing a reference camera image for a predefined reference camera pose;

[0017] - performing an image comparison between a reference camera image and the acquired camera image in order to identify pairs of feature points having feature points assigned to the same element and in order to determine the 2D image coordinates in the reference camera image and the 2D image coordinates in the acquired camera image;

[0018] - determining a camera pose change, in particular in the form of an essential matrix, based on the 2D image coordinates in the reference camera image and the 2D image coordinates in the acquired camera image; and

[0019] -Act the camera pose change on the reference camera pose to obtain camera pose data.

[0020] -Act the camera pose change on the reference camera pose to obtain camera pose data.

[0021] In order to display the display objects in the augmented reality data glasses in a contact-simulated manner, it is necessary to know the absolute glasses position relative to the reference system. If the data glasses are used in a vehicle, the reference system corresponds to the vehicle reference system, so that the display objects can be displayed in a contact-simulated manner relative to vehicle-fixed elements. In other words, the display objects are displayed superimposed at a specific location or superimposed on a specific element belonging to the vehicle.

[0022] In order to use data glasses to display display objects in a contact-simulated manner, it is necessary to know the accurate current eye posture data of the data glasses. The current eye posture data corresponds to the absolute eye posture data in the data glasses, which should describe the actual eye posture as accurately as possible and be used to display display objects in a contact-simulated manner.

[0023] The eyeglass pose includes the absolute position in space (3D position) and the 3D orientation along three spatial directions. In order to determine the pose, the data glasses are generally provided with an eyeglass motion sensing mechanism in the form of an IMU unit (Inertial Measurement Unit), which has an acceleration sensor and a gyroscope to determine the translation acceleration, angular acceleration or angular velocity as the eyeglass motion information. This information can be used from the absolute eyeglass pose data at a specific time point through a primary and / or secondary time integration, so as to re-determine the eyeglass pose data over time by accumulating incremental changes in the position and orientation of the data glasses.

[0024] In order to determine the change in the orientation of the data glasses, the angular velocity of the data glasses is integrated once over time. This allows the incremental change in orientation to be determined in order to redetermine the absolute glasses orientation. Compensation with vehicle movements detected in the vehicle is generally not necessary, since the orientation of the vehicle changes only at a very low frequency.

[0025] However, determining the change in position of the data glasses as part of the glasses pose is technically demanding, because in order to update the current absolute position of the current glasses pose data, the difference between the acceleration information of the glasses motion sensing mechanism and the translational acceleration information detected by the vehicle sensing mechanism fixed to the vehicle must be taken into account. In addition, kinematic terms in the moving vehicle must be taken into account, which kinematic terms include, for example, centripetal acceleration, Coriolis acceleration, etc. These terms cannot be measured directly and must be estimated by a tracking filter. In order to obtain the current change in position, the differential acceleration including the kinematic terms considered must first be integrated once over time to determine the speed of the data glasses, and then the speed must be re-integrated.

[0026] Due to the integration, when the absolute eyeglass pose data is recalculated, the drift increases quadratically with time, which may lead to large errors in the pose determination after a short time. Therefore, it is necessary to regularly correct the eyeglass pose data determined by the integration in order to compensate for this drift when determining the position and to be able to provide the current eyeglass position with sufficient accuracy.

[0027] Therefore, in order to provide current glasses pose data in the data glasses, it is necessary to regularly transmit absolute glasses pose data from the vehicle system outside the glasses (with respect to the vehicle coordinate system). This requires that the interior camera that records the camera image must be precisely known in terms of its position and orientation in the vehicle coordinate system, with which the data glasses in the vehicle interior can be identified with the aid of visual tracking methods and their glasses pose can be determined. Therefore, the camera image captured by the interior camera shows a section of the vehicle's interior that is clearly defined by orientation and position, i.e., camera pose, and the position of the data glasses identified in the camera image can therefore be precisely matched to the glasses position and glasses orientation with respect to the vehicle coordinate system.

[0028] The problem is that the orientation of the interior mirror can be changed by the user at any time, and the camera pose of the interior camera can therefore be uncertain.

[0029] To determine the camera pose, an initial calibration is provided, for example at the start of operation, by which reference camera images are defined for a known reference camera pose. The initial calibration can be fixedly predefined and in particular is derived from camera images actually acquired with a known camera pose.

[0030] In the reference camera image, some areas can be masked so that only non-masked areas of the interior view are used for comparison. In particular, the masked areas can correspond to areas of the interior view that show movable elements in the vehicle interior and / or that can be occupied by a vehicle occupant. Thus, by masking the reference camera image accordingly, areas in which elements can be dynamically adjusted during vehicle operation or areas in which a vehicle occupant can mask visible elements can be hidden.

[0031] To determine the camera pose, the recorded camera image can be compared with a reference camera image in order to find feature points that correspond to identical elements in the vehicle interior. These feature points can be identified by means of 2D image coordinates in the reference camera image and in the acquired camera image, so that features that correspond to one another in the reference camera image and in the acquired camera image can be determined using feature matching methods.

[0032] By comparing the reference camera image with the acquired camera image, identical feature points can be found, which do not need to be determined as reference points in advance. These feature points are generated by analyzing the acquired camera image and the reference camera image separately in order to obtain a certain number of feature points respectively. These feature points can be optimally matched by feature matching, for example with the aid of algorithms such as SIFT, SURF, ORB, AKZE, etc. The intrinsic matrix can be determined, for example, with the aid of RANSAC, from the feature point pairs. Then, the pose of the acquired camera image relative to the reference camera image is determined by the feature point pairs, the intrinsic matrix and the camera matrix.

[0033] The intrinsic matrix describes the epipolar geometry between a reference camera pose and the camera pose data to be determined, wherein the intrinsic parameters of the camera are known. The intrinsic matrix only includes a rotation matrix and a translation matrix.

[0034] The fundamental matrix describes the epipolar geometric relationship between the reference camera pose of the camera and the camera pose data to be determined. Unlike the essential matrix, the fundamental matrix takes into account both intrinsic and extrinsic camera parameters. If the intrinsic parameters of the camera are known, it is possible to transform from the fundamental matrix to the essential matrix.

[0035] The intrinsic matrix is ​​used to estimate the camera pose data (pose change) of the camera relative to a reference camera pose, and the intrinsic parameters of the camera are known.

[0036] The position change of the camera pose data of the interior space camera with respect to the reference camera pose can be ascertained from the 2D image coordinates of the feature points in the reference camera image and the 2D image coordinates of the feature points in the acquired camera image by means of algorithms known per se, such as SIFT, SURF, ORB, AKAZE, etc. This position change can be applied to the reference camera pose by a simple coordinate transformation in order to obtain the current camera pose data.

[0037] The reference camera image may correspond to a camera image recorded during an initial calibration process at a reference camera pose. Alternatively, the reference camera image may also be a camera image simulated at a preset reference camera pose, for example using Unity simulation, the camera image being created at a preset camera pose of an interior camera based on CAD data of the vehicle interior.

[0038] In particular, some areas in the reference camera image in which characteristic points are identified can be masked. The masked areas correspond to areas of the interior view that present movable elements in the vehicle interior and / or that can be covered by a vehicle occupant.

[0039] Thus, the interior view can be masked so that potential areas where movable elements may be present and areas where vehicle occupants may be located are not considered in the image comparison. It may therefore be sufficient to render the interior view only with elements of the vehicle interior that are fixed and visible when any seat is occupied.

[0040] According to a further aspect, a device, in particular a control unit, is specified for providing camera pose data, which describe the camera pose of an adjustable interior camera, in particular on an adjustable interior rearview mirror, wherein the device is designed to:

[0041] - Acquisition of camera images of the interior space camera;

[0042] - providing a reference camera image for a predefined reference camera pose;

[0043] - performing an image comparison between a reference camera image and the acquired camera image in order to identify pairs of feature points having feature points assigned to the same element and in order to determine the 2D image coordinates in the reference camera image and the 2D image coordinates in the acquired camera image;

[0044] - determining a camera pose change, in particular in the form of an essential matrix, based on the 2D image coordinates in the reference camera image and the 2D image coordinates in the acquired camera image; and

[0045] -Act the camera pose change on the reference camera pose to obtain camera pose data.

[0046] According to another aspect, a combination system for operating a vehicle system is provided, the combination system comprising:

[0047] - an adjustable interior rearview mirror having an interior camera arranged thereon / in it;

[0048] - the control unit as mentioned above;

[0049] - data glasses, the data glasses being configured to display at least one display object according to glasses posture data;

[0050] - wherein the control unit is further configured to determine absolute glasses pose data of the data glasses in the camera image based on the camera pose data by a visual tracking method and transmit the absolute glasses pose data to the data glasses. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The following is a detailed explanation of the implementation method with the help of the accompanying drawings. In the drawings:

[0052] Figure 1 A schematic top view of a vehicle system with a display system is shown;

[0053] Figure 2 A schematic diagram showing a display system with data glasses in a motor vehicle;

[0054] Figure 3 A flow chart is shown for illustrating a method for determining camera pose data of an interior camera. DETAILED DESCRIPTION

[0055] Figure 1 A schematic top view of a vehicle system with a vehicle 1 is shown, wherein a driver or vehicle occupant N wears data glasses 3 . The data glasses 3 are connected to a vehicle-mounted posture recognition system 2 via a communication connection 4 in a wired or wireless manner.

[0056] The communication connection 4 is designed as a data transmission channel, for example in the form of a wireless communication connection or a wired communication connection. The communication connection 4 can transmit any type of data and information between the assistance system 2 and the data glasses 3, for example based on data transmission packaged in data packets. The communication connection 4 can be based on Wi-Fi, Bluetooth, Bluetooth Low Energy or similar standardized radio protocols, for example.

[0057] Figure 2 A schematic diagram showing a vehicle system.

[0058] The assistance system 2 can be part of a vehicle assistance system and can in particular be arranged in a stationary manner in the vehicle 1 . The assistance system 2 can be equipped with a communication unit 23 which enables a communication connection 4 between the data glasses 3 and the assistance system 2 .

[0059] Furthermore, the assistance system 2 can be connected to a camera system 21, which has an interior camera 22 oriented toward the interior of the vehicle. The interior camera 22 is fixed to a movable interior rearview mirror, so that when the interior rearview mirror is adjusted or adjusted, for example, by the driver, the camera pose of the interior camera is adjusted. The camera system 21 can detect the vehicle occupant N as a camera image. The interior camera 22 can include, for example, an RGB camera, an IR camera, a fisheye camera, a dynamic vision sensor, etc.

[0060] The assistance system 2 can have a control unit 24 in order to analyze the camera image by means of a pattern recognition method and thereby determine absolute eyewear pose data of one or more data glasses 3 in the vehicle. The absolute eyewear pose data indicate the eyewear pose with respect to the vehicle coordinate system, i.e. the absolute eyewear position and the absolute eyewear orientation (preferably in all spatial directions in each case).

[0061] Furthermore, the assistance system 2 comprises a vehicle motion sensor system (IMU sensor system) 25, which can be designed, for example, in the form of a 6-DOF inertial sensor and provides motion data of the motion of the vehicle 1 in the form of translational acceleration or angular acceleration or angular velocity.

[0062] The data glasses 3 include two transparent viewing windows 32, which are embedded in a frame 31 in a manner known per se. The frame 31 is provided with glasses legs 33, for example, so that the data glasses 3 can be worn on the head of a user.

[0063] In addition, one or both viewing windows 32 (spectacles) are also provided with a transparent display surface 35, through which a display image for displaying a virtual display object can be projected into the eyes of the wearer of the data glasses 3 by a suitable device, for example, a display device 36 arranged on the frame 31. The display device 36 can have a microprocessor or a similar computing unit and a display unit, for example, a projection device or the like. The display unit can be configured to direct the electronically generated display image to the display surface 35 and image / display it there.

[0064] By means of the transparent design of the display surface 35, electronically generated images can be superimposed on the real environment that can be perceived through the display surface 35. Virtual display objects such as texts, symbols, video information, graphics, etc. can be presented on one or both display surfaces 35 by means of the display device 36.

[0065] The data glasses 3 can be worn on the user's head like a typical visual aid, wherein the data glasses 3 can be placed on the user's nose with the frame 31 and the legs 33 can be placed laterally on the user's head. The user's viewing direction in a linear direction is then achieved through the transparent display surface 35 of the viewing window 32, so that the user's viewing direction, which is predetermined by the eye position and the optical visual axis (eye axis), has a fixed relationship with the orientation of the data glasses 3. This relationship is individually related to the wearer of the data glasses 3 and is specified by the calibration information.

[0066] In order to display the display object, the corresponding object information in the form of object data is transmitted from the assistance system 2 to the data glasses 3 or is already provided in the data glasses 3. The object data specifies the type of display object, for example a text object, an icon or another identifier of a display area, one or more viewing areas in which the display object is to be displayed on the display surface 35, and the object position in the environment of the data glasses 3. The object position can be specified with respect to a world-fixed environment coordinate system or a vehicle-fixed vehicle coordinate system.

[0067] A glasses motion sensor mechanism 38 (IMU sensor mechanism, inertial sensor mechanism) may be provided, which is constructed, for example, in the form of a 6-DOF inertial sensor (acceleration sensor and gyroscope). It provides glasses motion information of the movement of the data glasses 3 in the form of translation acceleration and angular acceleration or angular velocity, and the glasses motion information can be converted into position change and orientation change, that is, differential glasses posture data, by corresponding (for example, quadratic) integration, so that the data for the glasses posture can be updated at a high frequency by using a time integration method.

[0068] The position changes and orientation changes (differential eyeglass pose) ascertained by the eyeglass motion sensor 38 can be used in a manner known per se to determine world-fixed current eyeglass pose data for the current eyeglass pose of the data eyeglasses 3. In order to ascertain vehicle-fixed eyeglass pose data, the eyeglass motion information must be corrected according to the vehicle motion information so that only the movement of the data eyeglasses 3 in the vehicle can be analyzed. In this case, the movement differences between the movement of the data eyeglasses and the movement of the vehicle are ascertained and analyzed to determine the current eyeglass pose data.

[0069] By means of the control unit 37, for example, object information can be received from the auxiliary system 2 via the communication device 39 or provided and processed in another way, so that the object information is displayed in a contact simulation in the corresponding viewing area according to the current glasses posture data about the environmental coordinate system, and the user of the data glasses 3 looks at the viewing area and can perceive the environmental objects in the area. In other words, when the environmental position of the display object is located in the viewing area, the display object given by the object information is displayed on the display surface 35.

[0070] The user N of the data glasses 3 is located in the vehicle 5 as an occupant, and the glasses posture of the data glasses 3 can be changed due to the movement of the head and the movement of the vehicle. In order to present the display object on the display surface of the data glasses 3 in a contact-simulated manner with respect to the environment position (i.e., the corresponding object information is equipped with the environment position), the current glasses posture data with respect to the environment coordinate system U must be present in the data glasses 3. Correspondingly, in order to present the display object on the display surface of the data glasses 3 in a contact-simulated manner with respect to the vehicle-fixed position (i.e., the corresponding object information is equipped with the position relative to the vehicle (vehicle-fixed)), the absolute glasses posture data with respect to the vehicle coordinate system F must be present in the data glasses 3.

[0071] In a display system, a method is implemented, such as, for example, according to Figure 3 The flowchart is described in detail.

[0072] In step S1 , a camera image is captured by the interior camera 22 .

[0073] The camera image shows a detail of the vehicle interior including the vehicle occupants, who are optionally wearing data glasses 3 .

[0074] Since the camera pose of the interior camera 22 may have changed due to the adjustment of the interior rearview mirror, the camera pose data are accordingly determined in step S2 by comparing a reference camera image (which was initially provided with known reference camera pose) with the acquired camera image. For the reference camera image, the initial reference camera pose is known in advance.

[0075] For the comparison between the reference camera image and the acquired camera image, feature points in the reference camera image and in the acquired camera image are first determined. Feature points that correspond to each other are determined by subsequent feature matching, thereby obtaining feature point pairs. For the feature points, the 2D image coordinates in the reference camera image and the 2D image coordinates in the acquired camera image are known.

[0076] By appropriately masking the initial reference camera image, areas of the reference camera image in which vehicle occupants may be located, for example on a vehicle seat, or movable elements of the vehicle interior, such as a backrest of a vehicle seat, etc., may be concealed for comparison.

[0077] The feature matching produces feature point pairs. The shift of the 2D image coordinates between the reference camera image and the acquired camera image enables the determination of the intrinsic matrix in step S3 based on known parameter estimation algorithms (e.g. RANSAC). From the intrinsic matrix and the reference camera pose (illustrated by, for example, the camera matrix), the camera pose change can be determined by means of matrix multiplication.

[0078] Subsequently, in step S4, the determined intrinsic matrix is ​​acted on the reference camera pose in order to obtain the current camera pose data. In this case, the position and orientation are determined in a manner known per se by scaling and matrix multiplication.

[0079] With the aid of the camera pose data and by carrying out a visual tracking method with the acquired camera images, the absolute eyewear pose data of the data eyewear 3 with respect to the vehicle coordinate system can be determined in step S5 by simple vector mathematics.

[0080] These absolute eyeglasses position data can now be transmitted to the data glasses 3 in step S6, so that in step S7 they can be provided in the data glasses 3 by means of a data fusion of the absolute eyeglasses position data newly determined by integration and the transmitted absolute eyeglasses position data. This can be used for a contact-simulated, i.e. positionally reliable, display of the display object.

[0081] The method described above can be carried out regularly, ie with the recording of each new camera image, or only after an adjustment of the interior mirror has been determined with the aid of other devices.

[0082] Reference numerals list

[0083] 1 Vehicle

[0084] 2Assistance systems

[0085] 3. Data Glasses

[0086] 4Communication Connection

[0087] 21-camera system

[0088] 22 cameras

[0089] 23 Communication Unit

[0090] 24Control Unit

[0091] 25 Vehicle motion sensing mechanism

[0092] 32 windows

[0093] 31 Frame

[0094] 33 glasses legs

[0095] 35 display surface

[0096] 36 Display device

[0097] 38 Eyeglasses motion sensing mechanism

[0098] 37 Control Unit

[0099] N users

Claims

1. A method for providing camera pose data, the camera pose data describing the camera pose of an adjustable interior space camera (22), the method comprising the following steps: - collecting (S1) a camera image of an interior space camera (22); - providing a reference camera image for a predefined reference camera pose; - performing (S2) an image comparison between the reference camera image and the acquired camera image in order to identify pairs of feature points having feature points assigned to the same element and in order to determine the 2D image coordinates in the reference camera image and the 2D image coordinates in the acquired camera image; - determining (S3) a camera pose change, in particular in the form of an essential matrix, based on the 2D image coordinates in the reference camera image and the 2D image coordinates in the acquired camera image; as well as -Act the camera pose change on the reference camera pose to obtain camera pose data.

2. The method according to claim 1, wherein: The method has the following further steps: - determining (S5) the absolute eyewear position of the data glasses (3) in the camera image by a visual tracking method based on the camera position data; - Transmitting (S6) the absolute eyewear position to the data eyewear (3).

3. The method according to claim 1 or 2, wherein: Regions are masked in the reference camera image so that only the uncovered regions of the interior view are used for comparison.

4. The method according to claim 3, wherein: The covered area corresponds to that area of ​​the interior view which shows movable elements in the vehicle interior and / or which can be occupied by a vehicle occupant.

5. The method according to any one of claims 1 to 4, wherein: The feature point pairs are determined with the aid of a feature matching method, in particular SIFT, SURF, ORB or AKAZE, wherein the camera pose change is determined from the feature point pairs with a parameter estimation method, in particular a RANSAC algorithm, and the feature point pairs respectively assign 2D image coordinates in a reference camera image to 2D image coordinates in the acquired camera image.

6. The method according to any one of claims 1 to 5, wherein: The reference camera image for a predefined reference camera pose is provided by predefining a reference camera pose and generating the reference camera image based on the reference camera pose by simulation based on CAD data of the vehicle interior, in particular using a graphics engine such as Unity.

7. A device, in particular a control unit (24), for providing camera pose data, which describe the camera pose of an adjustable interior camera (22), in particular on an adjustable interior rearview mirror, wherein: The device is configured to: - collecting camera images of the interior space camera (22); - providing a reference camera image for a predefined reference camera pose; - performing an image comparison between a reference camera image and the acquired camera image in order to identify pairs of feature points having feature points assigned to the same element and in order to determine the 2D image coordinates in the reference camera image and the 2D image coordinates in the acquired camera image; - determining a camera pose change, in particular in the form of an essential matrix, based on the 2D image coordinates in the reference camera image and the 2D image coordinates in the acquired camera image; as well as -Act the camera pose change on the reference camera pose to obtain camera pose data.

8. A combination system for operating a vehicle system, the combination system comprising: - an adjustable interior rearview mirror having an interior camera (22) arranged thereon / in it; - A control unit (24) according to claim 7; - data glasses (3), the data glasses being configured to display at least one display object according to the glasses position data; - wherein the control unit (24) is further configured to determine absolute glasses pose data of the data glasses (3) in the camera image by a visual tracking method based on the camera pose data, and transmit the absolute glasses pose data to the data glasses (3).

Images