Abstract: An image synthesizer for a surround monitoring system for a vehicle includes a receiving module for receiving a first image from a first camera, a second image from a second camera, and a control signal from a control module; and a combining module for combining the first image and the second image in the overlapping region and, depending on the control signal, to generate a combined image, the control signal indicating a condition external to the vehicle.

Abstract: A spherical image capture device may include two lenses, which may be protected by lens covers. During capture of visual content, usage of the lens covers may be checked to determine whether one or both of the lenses are covered by lens covers. Based on one or both of the lenses being covered by lens covers, an alarm may be generated to notify the user about the lens covers.

Abstract: A device may receive images of an object and information identifying the object, process the images using an artificial intelligence technique to identify parts of the object that are depicted in the images, and receive information identifying a location of damage on the object and information regarding the damage on the object. The device may process the information identifying the location of damage to identify a damaged part of the object, identify images depicting the damaged part, and identify, in the images, a location of the damaged part. The device may generate a first content item for display at the location of the damaged part in the images and generate a second content item for display with the images based on user interaction with the first content item, where the second content item includes information based on the information regarding the damage on the object.

Abstract: The present disclosure generally relates to streaming of media and, more specifically, to live streaming of media. The present disclosure relates, in particular, to live streaming of media for low-latency applications such as gaming applications (e.g., live casino gaming applications). The disclosure, among other things, describes a method of live streaming of media for low-latency applications, wherein the media streams are broadcasted (220) from a first device (e.g., a Media Streaming Provider Server System (140)) to one or several second devices (e.g., user game devices (110) for use in a live casino gaming application) over a WebRTC data channel.

Abstract: Cameras with large field-of-view lenses can cause significant geometric distortions of the images acquired. Training for object detection normally takes place on undistorted images. Thus, in order to detect objects of interest within the acquired images, an undistortion procedure is applied on the acquired images and an object detection is then performed on the undistorted images. Unfortunately, such undistortion procedure is too computationally expensive to be run on some edge devices. To remove the need to perform the undistortion procedure, it is proposed to train for object detection directly from distorted acquired images.

Abstract: An apparatus and method are provided for viewing panoramic images and videos through the selection of a particular viewing angle and window (zoom) within that panorama while allowing the viewer to simultaneously implement temporal transport control, allowing the video to be in a state of pause, play, fast forward, fast rewind, slow forward, slow rewind, or frame-by-frame. This capability may be used on video that is residing in memory on the viewer's viewing system, in a hard disk local to the viewer or in a shared location, or on a live buffered feed of video. A second capability of this apparatus and method relates to the use of a plurality of panoramic video or images from multiple synchronized cameras. In those cases, all panoramic video feeds are synchronized so that as a viewer pauses, rewinds, forwards a video in one panorama, all panoramas are time synchronized and go through the same states as the panorama being viewed.

Abstract: Multi-spectrum imaging systems and methods are provided for imaging in multiple spectrums, e.g., thermal IR (infrared) at wavelengths of 4 microns and 11 microns, near-IR, and visible light, all on a same optical centerline. For example, an imaging system includes a first imager and a second imager. The first imager includes an array of thermal IR detectors, wherein the first imager is configured to receive incident photonic radiation and generate a thermal IR image, wherein each thermal IR detector comprises a photon absorber member that is configured to absorb thermal IR photonic radiation from the incident photonic radiation, and reflect remaining photonic radiation in the incident photonic radiation along an optical path of the imaging system. The second imager is disposed in said optical path of the imaging system, wherein the second imager is configured to receive the remaining photonic radiation reflected from the first imager and generate a second image.

Abstract: An information processing device according to one embodiment of the present invention includes: a setting unit that sets a pattern used for outputting a partial image obtained from an omnidirectional image in accordance with the number and a position of at least one region of interest obtained from the omnidirectional image, which is a captured image captured by using a fisheye lens; an extract unit that, in accordance with a pattern set by the setting unit, cuts out a partial image including the region of interest from the omnidirectional image; a correction unit that performs a distortion correction process on the partial image cut out by the extract unit; and an output unit that, in accordance with the pattern set by the setting unit, outputs the partial image on which the distortion correction process has been performed by the correction unit.

Abstract: A vehicular surround vision system includes a plurality of vehicular cameras disposed at a vehicle and having respective exterior fields of view, a trailer camera disposed at a rear of a trailer and having a rearward field of view, an ECU and a display device in the vehicle. With the trailer not hitched to the vehicle and during a driving maneuver of the vehicle, the display device displays surround view video images derived from image data captured by at least the rearward viewing and sideward viewing vehicular cameras and provided to the ECU. With the trailer hitched to the vehicle and during a driving maneuver of the vehicle and trailer, the display device displays surround view video images derived from image data captured by the trailer camera and provided to the ECU and image data captured by at least some of the vehicular cameras and provided to the ECU.

Abstract: A marker for achieving more accurate position and posture estimation, and a posture estimation method and a position and posture estimation method using this marker are provided. In order to solve this problem, the present invention provides a marker for estimating the posture of an object, wherein the marker includes standard variable moire patterns VS1 and VS2 for estimating a visual line angle around at least one axis with respect to the object and high-sensitivity variable moire patterns VH1 and VH2 that have a resolution of visual line angle higher than the standard variable moire patterns VS1 and VS2 and that are used to estimate the visual line angle around the axis.

Abstract: An image display system includes an image acquisition unit which acquires an image, an area setting unit which sets a plurality of viewing areas to which the image is to be assigned according to a direction of the image, an image assignment unit which assigns the image to each of the plurality of viewing areas, an operation determination unit which discriminates an operation in a first viewing area among the plurality of viewing areas each having the image assigned thereto, and a display control unit which displays the image in the first viewing area and the image in a second viewing area other than the first viewing area among the plurality of viewing areas concurrently based on a result of the discrimination by the operation determination unit.

Abstract: An emergency stop apparatus includes a detecting device for detecting an obstacle intruding into a predetermined area around a working machine, an actuator fixed to a vehicle body of the construction machine or a member attached to the vehicle body, a coupling member for coupling an output shaft of the actuator with a brake pedal, and an electronic control unit. When the obstacle is detected by the detecting device, the electronic control unit actuates the actuator to move the brake pedal in an actuation direction of the service brake.

Abstract: An image capture device may switch operation between a spherical capture mode or a non-spherical capture mode. Operation of the image capture device in the spherical capture mode includes generation of spherical visual content based on the visual content generated by multiple image sensors. Operation of the image capture device in the non-spherical capture mode includes generation of non-spherical visual content based on visual content generated by a single image sensor.

Abstract: A method and system detects and localizes multiple instances of an object by first acquiring a frame of a three-dimensional (3D) scene with a sensor, and extracting features from the frame. The features are matched according to appearance similarity and triplets are formed among matching features. Based on 3D locations of the corresponding points in the matching triplets, a geometric transformation is computed. Matching triplets are clustered according to the computed geometric transformations. Since the set of features coining from two different object instances should have a single geometric transform, the output of clustering provides the features and poses of each object instance in the image.

Abstract: An image providing apparatus includes an internal communication interface configured to receive image data and camera setting information from a plurality of camera modules included in a network camera housing, and transmit a control signal to the plurality of camera modules; a processor configured to generate the control signal to control an operation of the plurality of camera modules, control a power supply to each of the plurality of camera modules, and generate combined image data by combining the image data obtained from the plurality of camera modules; and an external communication interface configured to transmit the combined image data to a client terminal via an Internet network.

Abstract: A method, system, apparatus, and/or device for sensing determining levels of an object to execute a command. The method, system, apparatus, and/or device may include: sensing an object that occupies a first portion of a field of view (FOV) of the sensor at a first point in time; determining a first saturation level of the object at the first portion of the FOV; sensing the object that occupies a second portion of the FOV of the sensor at a second point in time; determining a second saturation level of the object at the second portion of the FOV; determining that the first saturation level is different than the second saturation level; and in response to the first saturation level being different than the second saturation level, executing an executable command associated with the first saturation level and the second saturation level.

Abstract: Dual-lens assemblies and cameras including dual lens-assemblies that include a first lens barrel securing a first lens having a first optical axis and a second lens barrel securing a second lens having a second optical axis are disclosed. In one dual-lens assembly, the first optical axis is approximately parallel to and spaced from the second optical axis by a lateral offset, axial lengths of the first lens barrel and the second lens barrel are approximately equal, and the first lens and the second lens are oriented in opposite directions at opposing ends of the first lens barrel and the second lens barrel.

Abstract: A method of decoding a coded video sequence in a decoder, into a series of reconstructed pictures, is provided, the reconstructed pictures representing a projection on a planar surface of a non-planar surface. The method includes generating an ordered candidate list including a plurality of candidate motion vectors for a current prediction unit (PU) under reconstruction, by ordering the plurality of candidate motion vectors in the ordered candidate list based on a geometric property of the projection including the current PU and based on a position of the current PU in the projection, decoding an index, referencing the generated ordered candidate list using the decoded index to select one of the plurality of candidate motion vectors from the generated ordered candidate list, generating a prediction signal, using the selected one of the plurality of candidate motion vectors, and reconstructing the current PU, using the generated prediction signal.

Abstract: An information processing apparatus includes circuitry to register at least a first viewpoint and a second viewpoint designated next to the first viewpoint as viewpoints in a full-view spherical image with an order designated by a user, set a transition path of the viewpoints from the first viewpoint to the second viewpoint by interpolating between the first viewpoint and the second viewpoint in the full-view spherical image, generate a first partial image having a center that matches the first viewpoint, and a second partial image having a center that matches the second viewpoint, and play animation by sequentially displaying the first partial image and the second partial image with the order designated by the user while transiting the viewpoints from the first viewpoint to the second viewpoint along the set transition path.

Abstract: A panoramic camera is provided with an orientation sensor to record orientation along with a panoramic image. The display or print image is selected from the captured panoramic image, a warped, non-scale-correct image. The selection of the display image is done using the orientation sensor. The orientation sensor data can be recorded with the panoramic image as metadata. The orientation sensor can also be used to control an external display and/or panoramic image selection during browsing.

Abstract: The present inventive concept relates to a kiosk design for an advanced photographic system. More specifically, the present inventive concept relates to a self-contained, automated photobooth kiosk. In embodiments of the present inventive concept, the photobooth kiosk is capable of taking a 360 degree panoramic photograph or sequence of photographs of a subject and surrounding background. For instance, a customer of the photobooth kiosk may stand in the center of the photobooth and have his or her picture taken with a plurality of specialized “machine vision” cameras, with the images sent to a central processor such as a computer for processing into a 360 degree panoramic photograph or video clip. After the photo-taking session, the customer may collect prints of the pictures at the kiosk, similar to presently available photobooths.

Abstract: Embodiments are generally directed to selective packing of patches for immersive video. An embodiment of a processing system includes one or more processor cores; and a memory to store data for immersive video, the data including a plurality of patches for multiple projection directions. The system is select the patches for packing, the selection of the patches based at least in part on which of the multiple projection directions is associated with each of the patches. The system is to encode the patches into one or more coded pictures according to the selection of the patches.

Abstract: The present invention discloses a method for performing original real-time video broadcasting with non-stationary panoramic video on virtual reality headset, comprising PC terminal one and PC terminal two, also a panoramic camera connected to the PC terminal one, a four-axis electromechanical structure connected to the panoramic camera, and a virtual reality headset connected to the PC terminal two. The present invention enables the panoramic camera to perform live video transmission on the virtual reality headset in low delay or real time, which will realize virtual reality of different scenarios, for example, the panoramic HD camera can be placed in the car cockpit, restoring the driving environment for the user perfectly. With real-time panoramic video transmission, user may complete the remote control of the car with the same experience.

Abstract: Implementations generally relate to producing a still image from a video or series of continuous frames. In some implementations, a method includes receiving the frames that a capture device shot while moving in at least two dimensions. The method further includes analyzing the frames to determine changes of positions of objects in at least two of the frames due to movement of the objects in the scene relative to changes of positions of objects due to the movement of the capture device during the shoot time. The method further includes determining, based at least in part on the variability of the objects, one or more target subjects which the capture device captures during the shoot time. One or more still images are generated from the plurality of frames having at least a portion of the target subject.

Abstract: A method for generating a three-dimensional (3D) lane model, the method including calculating a free space indicating a driving-allowed area based on a driving image captured from a vehicle camera, generating a dominant plane indicating plane information of a road based on either or both of depth information of the free space and a depth map corresponding to a front of the vehicle, and generating a 3D short-distance road model based on the dominant plane.

Abstract: An image monitoring system includes: recording means for recording an image captured by a camera via a network; control means for controlling the system so as to display the present image captured by the camera or a past image recorded on the recording means on display means; and moving-object detecting means for detecting a moving object from the image captured by the camera; wherein the moving-object detecting means includes resolution conversion means for generating an image with a resolution lower than the resolution of the image captured by the camera, positional-information output means for detecting a moving object from the image generated by the resolution conversion means and outputting positional information on the detected moving object, and information merging means for merging the positional information of the moving object with the image captured by the camera on the basis of the positional information of the moving object output by the positional-information output means.

Abstract: A system for generating a 360-degree viewing experience may receive a plurality of images of an object from an image capture device, wherein each of the plurality of images corresponds to a different rotational orientation of the object relative to the image capture device. The system may detect, using a first machine learning model, the object in each of the plurality of images. The system may detect, using a second machine learning model, regions associated with identifiable object features in one or more images of the plurality of images. The system may assign feature metadata to the one or more images, the features metadata associated with one or more detected regions of the detected regions of the object in the one or more images. The system may publish, with an application programming interface, the plurality of images and the feature metadata for the 360-degree viewing experience.

Abstract: Various technologies described herein pertain to rectification of a fisheye image. A computing system receives the fisheye image. Responsive to receiving the fisheye image, the computing system applies a first lookup function to a first portion of the fisheye image to mitigate spatial distortions of the fisheye image. The computing system also applies a second lookup function to a second portion of the fisheye image to mitigate the spatial distortions. The first lookup function maps first pixels in the first portion to a first rectilinear image corresponding to the first portion when viewed from a first perspective of a first virtual camera. The second lookup function maps second pixels in the second portion to a second rectilinear image corresponding to the second portion when viewed from a second perspective of a second virtual camera. The computing system then outputs the first rectilinear image and the second rectilinear image.

Abstract: A vehicle exterior environment recognition apparatus includes first and luminance-image acquiring units, first and second distance-image generating units, and a composite-image generating unit. The first luminance-image acquiring unit acquires first luminance images captured by image-capturing units that are disposed at positions different from each other at a first exposure time. The second luminance-image acquiring unit acquires second luminance images captured by the image-capturing units at a second exposure time shorter than the first exposure time. The first distance-image generating unit generates a first distance image through pattern matching of the first luminance images. The second distance-image generating unit generates a second distance image through pattern matching of the second luminance images.

Abstract: A controller and method therein for controlling encoding of a set of images to enable blending of an overlapping area, where a first image and a second image overlap each other are disclosed. The controller encodes macroblocks of the non-overlapping area in the first image using a set of base quantization parameter values, QP-values, and adds the same set of base QP-values to a header of each macroblock. The controller encodes macroblocks of the overlapping area in the first image using a set of first QP values, and adds a modified set of the first QP-values to a header of each macroblock. The controller encodes macroblocks of the overlapping area in the second image using a set of second QP values, and adds a modified set of the second OP-values to a header of each macroblock.

Abstract: Provided are systems and methods for processing 360-degree video data. In various implementations, a spherical representation of a 360-degree video frame can be segmented into a top region, a bottom region, and a middle region. Using a cylindrical equal area projection, such as the Lambert cylindrical equal area projection, the middle region can be mapped into one or more rectangular areas of an output video frame.

Abstract: An apparatus for three-dimensional inspection includes a carrier, an image sensing component, and a processor. The carrier is configured to hold an object. The image sensing component is configured to capture a first image, a second image and a third image of the object along a first axis, a second axis, and a third axis respectively, and the first axis, the second axis, and the third axis are not parallel with each other. The processor is configured to analyze the first image and the second image to obtain a first directional stereo information, and analyze the third image and a determined image of the object to obtain a second directional stereo information.

Abstract: A method and an apparatus for a terminal, where the method includes obtaining a first video sequence photographed by a first camera and a second video sequence photographed by a second camera, determining, according to a first image in the first video sequence and photographed at a first moment and a second image in the second video sequence and photographed at the first moment, an available first motion vector (V), and a third image and a fourth image corresponding to the V, obtaining a second motion vector (U) of the first image relative to the third image, obtaining a third motion vector (W) of the fourth image relative to the second image, and obtaining a correspondence between a pixel in the first image and a pixel in the second image according to the V, the U, and the W.

Abstract: An image processing apparatus detecting a plurality of joint positions between a plurality of input images includes a target image generation unit configured to generate a plurality of target images to be searched in a second input image among the input images from a first input image among the input images; a characteristic amount calculation unit configured to calculate a characteristic amount for each of the target images generated by the target image generation unit; and a joint position determination unit configured to determine the joint position for a target image of interest among the target images having the characteristic amounts calculated by the characteristic amount calculation unit depending on the characteristic amount.

Abstract: A surround view system for a vehicle includes a processor (10) and cameras (2 to 5) that can be arranged on the vehicle so that the cameras (2 to 5) can record images of an outside environment of the vehicle. The processor is configured to analyze a position (?) and/or a movement of a movable part (14) of the vehicle (1) and to generate a composite image of the outside environment from individual images recorded by the cameras. Furthermore, the processor is configured to calculate an image processing region within the composite image for an adaptive image processing, to determine, based on the analyzed position and/or movement of the movable part (14), that a back-projection of the movable part (14) goes beyond the image processing region, and to modify the image processing region so that the back-projection of the movable part (14) is within the modified image processing region.

Abstract: A tiltable system for panoramic imaging of a scene.

Abstract: A system captures a first hemispherical image and a second hemispherical image, each hemispherical image including an overlap portion, the overlap portions capturing a same field of view, the two hemispherical images collectively comprising a spherical FOV and separated along a longitudinal plane. The system maps a modified first hemispherical image to a first portion of the 2D projection of a cubic image, the modified first hemispherical image including a non-overlap portion of the first hemispherical image, and maps a modified second hemispherical image to a second portion of the 2D projection of the cubic image, the modified second hemispherical image also including a non-overlap portion. The system maps the overlap portions of the first hemispherical image and the second hemispherical image to the 2D projection of the cubic image, and encodes the 2D projection of the cubic image to generate an encoded image representative of the spherical FOV.

Abstract: An information processing apparatus, an information processing system, a data generation method, and a recording medium storing program code. The information processing apparatus and data generation method include displaying a spherical image obtained by an imaging apparatus, reproducing three-dimensional audio data obtained by the imaging apparatus in association with the spherical image, accepting designation of a displaying direction on the spherical image, generating two-dimensional image data that corresponds to the displaying direction, based on data of the spherical image, and generating two-dimensional audio data that corresponds to the displaying direction, from the three-dimensional audio data. The information processing system includes the information processing, and an image processing server connected to the information processing apparatus through a communication network.

Abstract: A panoramic imaging system includes a plurality of separated video cameras that may be distributed around an object. A series of images captured by at least one of the separated video cameras is stored in a predetermined file format. The panoramic system further includes a viewer module that may render the series of images using the first file format. Moreover, the panoramic system includes a calibration module capable of modifying information associated with at least one of the series of images, where the modification results in calibration data stored in a configuration file and/or in the predetermined file format. The viewer module may also be capable of rendering the series of images in accordance with the predetermined file format.

Abstract: A video processing method includes receiving a virtual reality (VR) content, encoding visual data obtained from the VR content to generate a part of a coded bitstream, and encapsulating the part of the coded bitstream into ISO Base Media File Format (ISOBMFF) file (s). In one exemplary implementation, the ISOBMFF file (s) may include a timed metadata track associated with a sphere visual track, where the timed metadata track is set to signal that the associated sphere visual track contains at least one spherical region contributed from at least one region visual track. In another exemplary implementation, the ISOBMFF file (s) may include a timed metadata track associated with a region visual track, where the timed metadata track is set to signal that the associated region visual track contributes to at least one spherical region carried in at least one sphere visual track. Further, an associated video processing apparatus is provided.

Abstract: A method, computer program product and system for eye contact-based information transfer. Communication is established by a computer device between a first head mounted device worn by a first user and a second head mounted device worn by a second user. Gaze tracking establishes through gaze tracking of the first head mounted device and the second head mounted device that the first head mounted device and the second head mounted device have made mutual eye contact. Responsive to an input to the first head mounted device made while the established mutual eye contact exists, transferring, by the computer device, a file from the first head mounted device to the second head mounted device.

Abstract: An electronic device is provided. The electronic device includes a display, and a processor functionally connected to the display and configured to, when a first input is received while a first partial image, which corresponds to a first direction, in an image taken by at least one camera is displayed on the display, control the display to display a second partial image, which corresponds to a second direction different from the first direction, in the image in response to the first input.

Abstract: A multiple-wavelength image analysis electro-optical system for detecting a disabled ship and persons overboard of the present invention is a multiple-wavelength image analysis electro-optical system for detecting a disabled ship and persons overboard configured to have an input part 10 provided with an ultra-low light camera 3, a short-wavelength infrared image sensor 5, a medium-wavelength infrared image sensor 7, and a long-wavelength infrared image sensor 9; a signal processing part 20 for receiving and processing data of the input part 10; a display part 30 for receiving and displaying data of the signal processing part 20; a storage part 40 for storing data of the signal processing part 20 and the display part 30; and a control part 90 provided with a camera control board 50 and a drive control board 60 for controlling the input part 10, signal processing part 20, the display part 30 and the storage part 40.

Abstract: According to a control device of the present invention, there is provided a center server including a display processing unit that performs control so as to display a panorama image display area in which a panorama image captured with a camera terminal device is displayed and an enlarged image display area in which an enlarged image is displayed, the enlarged image being obtained by enlarging an area corresponding to a position specified on the panorama image, and a control unit that performs control so as to capture with the camera terminal device an image of the area corresponding to the position specified on the panorama image. The display processing unit displays on the enlarged image display area the image captured with the camera terminal device in accordance with control by the control unit, and the panorama image is a circular panorama image.

Abstract: Illuminated signal devices and speed detectors for audio/video (A/V) recording and communication devices in accordance with various embodiments of the present disclosure are provided. In one embodiment, an illuminated signal device configured for capturing image data is provided, the device comprising a camera having a field of view, a communication module, and a processing module operatively connected to the camera and the communication module, the processing module comprising a processor, and a signal device application, wherein the signal device application configures the processor to detect motion within the field of view of the camera, capture image data in response to the detected motion using the camera, and transmit the image data to a backend server using the communication module.

Abstract: Disclosed is a panoramic image mapping method and a corresponding reversely mapping method. Particularly, the mapping process includes mapping a panoramic image or a spherical surface corresponding to Video A: first, dividing the spherical surface into three areas based on the latitudes of the spherical surface, denoted as Area I, Area II, and Area III, respectively; mapping the three areas to a square plane I?, a rectangular plane II?, and a square plane III?, respectively; then, splicing the planes I?, II? and III? into a plane, wherein the resulting plane is the two-dimensional image or video B. Compared with the equirectangular mapping method, the method according to the present disclosure may effectively ameliorate oversampling in high-latitude areas and effectively lower the bit rate needed by coding and the complexity of decoding. The present disclosure relates to the field of virtual reality, which may be applied to panoramic images and videos.

Abstract: A monitoring device includes: a communication interface configured to receive image data acquired by a camera and a signal generated by at least one sensor; a screen configured to display thereon the received image data as a main image; and a controller configured to control operations of the communication interface and the screen, wherein the controller controls the screen to include a control area around an area for displaying the main image so that the control area displays the main image and a surrounding image of the main image, and wherein, in response to the sensor sensing at least one event, the controller controls generation of an alarm with indication of at least one pointer associated with the sensor in the control area as a source of the sensing the event, and controls the main image to be changed to cover an area where the sensor is disposed.

Abstract: There is described a system for providing streaming services, comprising: a plurality of capture devices each for generating a stream of an event, each stream being associated with a timing reference; and a server for analysing the plurality of captured streams in order to align the received plurality of received captured streams according.

Abstract: An image provision device (100) displays on a field-of-vision covering display (192) an image obtained by trimming all-surroundings image data (21). A detector (11) detects an orientation or a change in the orientation of a user's head or sight line by means of a sensor (191). A direction setter (12) sets an eye direction to the all-surroundings image data (21) in accordance with the detected orientation or change in the orientation. An extractor (13) obtains a zoomed image by trimming and zooming part of the all-surroundings image with a zoom factor that depends on an angle formed between the set eye direction and a predetermined direction in the all-surroundings image data (21). A displayer (14) displays the obtained zoomed image on a display (192).

Abstract: Image processing device for producing in real-time a digital composite image from a sequence of digital images recorded by a camera device, in particular an endoscopic camera device, the image processing device including a selecting unit, a key point detection unit, a transforming unit and a joining unit, wherein the key point detection unit includes a maximum detection unit configured for executing following steps separately for the filter response for the reference image and for the filter response for the further image, wherein a variable threshold is used: i) creating blocks by dividing the respective filter response, ii) calculating the variable threshold for each of the blocks, iii) discarding those blocks of the blocks from further consideration, in which the respective filter response at a reference point of the respective block is less than the respective variable threshold.