Abstract: Systems and methods for calibrating a sensor array for 3D depth sensing include the steps of providing multiple 3D sensors each for (i) illuminating a field of view of the sensor and (ii) generating an output array of pixelwise values indicative of distances to objects within the illuminated a field of view; sequentially causing each of the 3D sensors to generate an output array while other 3D sensors are illuminating their fields of view; and creating an interference matrix from the generated output arrays, the interference matrix indicating, for each of the 3D sensors, a degree of interference by other 3D sensors simultaneously active therewith.

Abstract: A monitoring system includes a plurality of entrance cameras that capture each of visitors from multiple different angles when the each of visitors enters a facility in which a plurality of intra-facility cameras are installed at respective prescribed positions, and a monitoring apparatus that accumulates visitor data which associates captured images of the each of visitors with identification information of the each of visitors, and monitors the each of visitors in the facility. The monitoring apparatus specifies one intra-facility camera installed at or near a position of detection of the abnormality as a near-abnormality-position intra-facility camera in response to a notice of detection of an abnormality in the facility, extracts one or more visitors found in a video captured by the near-abnormality-position intra-facility camera based on the visitor data; and employs the one or more visitors as security action target persons.

Abstract: A stereo vision measuring system and measuring method are provided. The system includes a main control device and at least two cooperation devices; each of the cooperation devices is a movable device, and is provided with an image acquiring part; the cooperation device is configured to acquire a first image of a first scene and first image information of the first image and transmit them to the main control device in a calibrating phase, and the cooperation device is further configured to acquire a second image of a second scene and second image information of the second image, and transmit the second image information to the main control device in a measuring phase; the main control device is configured to acquire parameter information of the image acquiring parts in the calibrating phase; the main control device is further configured to obtain three-dimensional reconstruction information of the second scene.

Abstract: Provided are a three-dimensional image capturing device and a method. The device includes: a first lens module and a second lens module, and an adjustable mechanism connecting the first lens module and the second lens module respectively. The adjustable mechanism is capable of changing a lens interval between the first lens module and the second lens module; and a processing apparatus connected to the first lens module and the second lens module, configured to determine a capturing interval between the three-dimensional image capturing device and a target capturing object, determine a target lens interval corresponding to the capturing interval; control the adjustable mechanism to adjust an interval between the first lens module and second lens module to be the target lens interval, and capture a three-dimensional of the target capturing object by using first lens module and second lens module, which enlarges the capturing range of three-dimensional image capturing device.

Abstract: An active illumination range camera comprising illumination and imaging systems that is operable to provide a range image of a scene in the imaging system's field of view (FOV) by partitioning the range camera FOV into sub-FOVs, and controlling the illumination and imaging systems to sequentially illuminate and image portions of the scene located in the respective sub-FOVs.

Abstract: System and techniques for vehicle environment modeling with a camera are described herein. A time-ordered sequence of images representative of a road surface may be obtained. An image from this sequence is a current image. A data set may then be provided to an artificial neural network (ANN) to produce a three-dimensional structure of a scene. Here, the data set includes a portion of the sequence of images that includes the current image, motion of the sensor from which the images were obtained, and an epipole. The road surface is then modeled using the three-dimensional structure of the scene.

Abstract: A head-mounted display is disclosed. The head-mounted display includes: a housing adapted to be worn on a user's head, the housing defining a partially enclosed chamber which covers the user's eyes when the housing is worn by the user; at least one display unit mounted in the chamber; a processor coupled to the at least one display unit; and a humidifier coupled to the housing, the humidifier being configured to controllably increase moisture in the chamber.

Abstract: Disclosed is a display apparatus for vehicle including: a display unit configured to include a barrier unit and an image output unit; and a processor configured to output a parallax barrier 3D image to the display unit by implementing a parallax barrier in the barrier unit and outputting a disparity image to the image output unit, and to adjust an open slit ratio of the barrier unit based on an average luminance value of the disparity image.

Abstract: A vicinity supervising device includes an image capturing unit that captures a plurality of images in which an object is simultaneously captured from different locations; a matching processing unit that calculates a parallax point and a reliability point, the parallax point being the number of pixel regions from which a parallax is calculated among a plurality of pixel regions divided by the plurality of images, and the reliability point being the number of pixel regions having high recognition reliability among the plurality of pixel regions; and an evaluation unit that evaluates, based on the parallax point and the reliability point, an imaging environment where the plurality of images are captured.

Abstract: A system and method multi-camera error compensation including recording a plurality of raw images via a plurality of digital cameras and an application processor processing with a multi-view stereo imaging system one or more plural image frames from the raw images captured by the plurality of digital cameras. The plural images may be stored in memory and the detection conducted for calibration loss of at least one digital camera via the processor executing instructions for a multi-camera error compensating system to determine loss of calibration in plural images. The multi-camera error compensating system conducts dynamic recalibration of plural image calibration parameters based on at least one plural image frame and in response to detection of calibration loss via the multi-camera error compensating system and a processor reprocesses the plural image frame from a reprocessing queue according to the recalibrated plural image parameters.

Abstract: Provided is a method of acquiring depth information of an object including acquiring first depth information of the object located at a first point in a common sensing region common to respective sensing regions of a plurality of sensors and first shape information of at least one view point, acquiring second shape information of the object at a second point when the object is located at the second point in a region excluding the common sensing region on the respective sensing regions of the plurality of sensors, and acquiring second depth information of the object at the second point based on a result of comparison between the first shape information of the at least one view point and the second shape information and the first depth information.

Abstract: An apparatus includes a first camera module providing a first image of an object with a first field of view, a second camera module providing a second image of the object with a second field of view different from the first field of view, a first depth map generator that generates a first depth map of the first image based on the first image and the second image, and a second depth map generator that generates a second depth map of the second image based on the first image, the second image, and the first depth map.

Abstract: The present invention comprises an apparatus and method for monitoring a zone, which is a contiguous or non-contiguous, two-dimensional (2D) area or three-dimensional (3D) volume. Among its several advantages, the apparatus includes a plurality of sensors that monitor portions of the zone and report intrusion status to a control unit that provides monitoring boundary information to each of the sensors based on user input and further “maps” or otherwise associates each sensor to control unit outputs in accordance with user-defined behaviors. Still further, as a meaningful aid for configuring monitoring boundaries used by the sensors for objection intrusion detection, in one or more embodiments of the apparatus and method, at least a subset of sensors use a common coordinate frame of reference, based on a common origin located within overlapping sensor fields of view.

Abstract: A method of generating a look-up table, LUT, for constructing a dewarped B-scan image from A-scans of a cropped part of a sequence of acquired OCT A-scans, the LUT associating each of a plurality of pixel arrays that are to form the dewarped B-scan image with a respective A-scan in the cropped part. The method comprises: using an indication of a spatial distribution of scan locations of A-scans to determine a dewarp function for selecting, from among acquired A-scans, A-scans having uniformly spaced scan locations; using the function and cropping information, in accordance with which the cropped part is cropped from the acquired sequence, to select, for each array, a respective A-scan from the sequence such that A-scans are selected from the cropped part and have uniformly spaced scan locations; and storing, for each array, a respective pixel array identifier in association with a respective identifier of the selected A-scan.

Abstract: A human body modelling method and apparatus and an electronic device. The method comprises: acquiring at least one human body image; extracting an original profile of a human body from the human body image; calculating a body surface profile of the human body according to the original profile; and constructing a three dimensional human body model according to the body surface profile. The establishment of a three dimensional human body model originates from two dimensional human body images, without using a human body scanner to scan a human body, which reduces the cost of establishing the three dimensional human body model. Moreover, there is no requirement for the position and dressing of a human body object, and the process of establishing the three dimensional human body model is more convenient.

Abstract: Image processing can include storing a first image of a scene captured by a first camera, storing a second image of the scene captured by a second camera, and storing a first user-composed image. The first user-composed image can be derived from at least one of: (a) the first image and (b) the second image. The image processing can further include: receiving a second command to edit the first user-composed image, editing the first user-composed image, and saving the edited first user-composed image as a second user-composed image.

Abstract: The present application discloses a combinatory sensor apparatus. The apparatus includes a sensor mounting frame and at least one sensor; the sensor mounting frame includes a sensor mounting base plate and at least one sensor support component, the sensor mounting base plate is mounted on a top of a driverless vehicle, and the at least one sensor support component is mounted on the sensor mounting base plate for supporting the at least one sensor; and the at least one sensor includes a global positioning system antenna and/or a laser radar. This implementation guarantees the consistency of sensor installation locations.

Abstract: Various embodiments disclosed herein include techniques for maintaining multiple cameras in focus on same objects and/or at same distances. In some examples, a subordinate camera may be configured to focus based on the focus of a primary camera. For instance, a focus relationship between the primary camera and the subordinate camera may be determined. The focus relationship may characterize the trajectory of the lens position of the subordinate camera with respect to the lens position of the primary camera. In various examples, the focus relationship may be updated.

Abstract: The present invention describes a system for calibrating a plurality of cameras in an area. The system functions by using moving objects to calibrate, in particular using people. In addition, the system implements automatic re-calibration in a specific way to reduce human intervention, cost and time.

Abstract: Methods, devices and systems are disclosed for improved depth perception in stereoscopic night vision devices. Among these are embodiments for aligning information overlays in the stereo view with associated objects, and for generating stereo information from single lenses or intensifiers. In some illustrative embodiments, a camera and position sensor are provided for at least two viewers, e.g., a pilot and a copilot, such that when a scene overlaps between viewers, the system produces a stereoptic scene, in which the users can more accurately determine a difference in depth between two or more distant objects. An illustrative binocular night vision system uses a high-resolution depth map to present binocular images to a user. In some embodiments, supplementary content can be overlaid, with an appropriate binocular disparity that is based on the depth map.

Abstract: A special-purpose hardware device for mitigating motion-to-photon latency in head-mounted displays may include an image signal processor that receives at least one image frame captured by a camera device of a head-mounted-display system. The special-purpose hardware device may also include an input-formatting component that receives the computer-generated imagery. The special-purpose hardware device may further include a blending component that generates at least one mixed-reality frame by overlaying the computer-generated imagery onto the image frame. The special-purpose hardware device may additionally include a frame-output interface that feeds the mixed-reality frame generated by the blending component to a display device of the head-mounted-display system to facilitate displaying the mixed-reality frame for presentation to a user wearing the head-mounted-display system. Various other apparatuses, systems, and methods are also disclosed.

Abstract: A pair of cameras having an overlapping field of view is aligned based on images captured by image sensors of the pair of cameras. A pixel shift is identified between the images. Based on the identified pixel shift, a calibration is applied to one or both of the pair of cameras. To determine the pixel shift, the camera applies correlation methods including edge matching. Calibrating the pair of cameras may include adjusting a read window on an image sensor. The pixel shift can also be used to determine a time lag, which can be used to synchronize subsequent image captures.

Abstract: Systems and methods are disclosed whereby a surface detection system is employed to obtain intraoperative surface data characterizing an exposed surface of the spine. In some embodiments, this intraoperative surface data is registered to segmented surface data obtained from volumetric data of the spine in order to assess the intraoperative orientation of the spine and provide feedback associated with the intraoperative orientation of the spine. The feedback may characterize the intraoperative spinal orientation as a change relative to the preoperative orientation.

Abstract: Provided is a three-dimensional reconstruction method of reconstructing a three-dimensional model from multi-view images. The method includes: selecting two frames from the multi-view images; calculating image information of each of the two frames; selecting a method of calculating corresponding keypoints in the two frames, according to the image information; and calculating the corresponding keypoints using the method of calculating corresponding keypoints selected in the selecting of the method of calculating corresponding keypoints.

Abstract: A 3D sensing device includes a base, an infrared emitting module, a first infrared receiver, a second infrared receiver, a color camera module, a first moving member, a second moving member, a first driving member, and a second driving member. The first moving member and the second moving member are slidably mounted on the base. The first driving member and the second riving member respectively drive the first moving member and the second moving member to slide along the base. The infrared emitting module is mounted on the first moving member. The color camera module is located between the first infrared receiver and the infrared emitter. The second infrared receiver is mounted on the second moving member. The first moving member and the second moving member respectively move the infrared emitting module and the second infrared receiver along the base relative to the first infrared receiver.

Abstract: An apparatus for use in visual training of a subject, the apparatus including an at least partially opaque elongate occluder for at least partially occluding a substantially laterally oriented area of a field of view of a first eye so as to define an occluded area and a non-occluded area, and, wherein in use the occluded area moves relative to the field of view, and the subject views: at least one target that extends into the occluded area using both eyes so that the subject perceives an image of the target, the image in the non-occluded area being at least partially based on signals from the first eye and the image in the occluded area being at least partially based on signals from a second eye; and, ensures the target is at least partially aligned in the occluded and non-occluded areas.

Abstract: Obtaining one or more parameters of an autonomous vehicle, the parameters including any of a position, path, and/or speed of the autonomous vehicle. A region of interest from a plurality of regions surrounding the autonomous vehicle is identified based on the one or more parameters. One or more sensors mounted on a sensor guide rail are controlled, based on the region of interest, to move the sensor(s) along at least a portion of the autonomous vehicle, and to capture sensor data of the region of interest and not capture sensor data from the one or more other regions of the plurality of regions surrounding the autonomous vehicle, the sensor guide rail being mounted on a surface of the autonomous vehicle. The captured sensor data is provided to a processor capable of facilitating, based on the captured sensor data of the region of interest, one or more autonomous vehicle driving actions.

Abstract: In an imaging apparatus, a pupil division optical system includes a first pupil transmitting light of a first wavelength band and a second pupil transmitting light of a second wavelength band different from the first wavelength band. An imaging device is configured to capture an image of light transmitted through the pupil division optical system and a first color filter having a first transmittance characteristic and light transmitted through the pupil division optical system and a second color filter having a second transmittance characteristic partially overlapping the first transmittance characteristic, and output the captured image. A processor is configured to generate a monochrome correction image or a correction image by correcting a value that is based on components overlapping between the first transmittance characteristic and the second transmittance characteristic for the captured image.

Abstract: The present disclosure provides a robot-based 3D picture shooting method and system, and a robot using the same. The method includes: obtaining a distance between a photographed object and the photographing device of the robot based on a received shooting instruction; calculating an inter-axis distance based on the distance; obtaining the first picture after moving the robot for half of the inter-axis distance along the movement direction; obtaining the second picture after moving the robot for entire of the inter-axis distance from a current position along an opposite direction of the movement direction; and synthesizing the first picture and the second picture to obtain a 3D picture of the photographed object. In the process, the robot moves the photographing device according to the calculated inter-axis distance, and obtains two pictures of the left and right of the photographed object, which is not necessary to use a binocular camera.

Abstract: A monitoring system includes a plurality of cameras and a control device. The control device includes device storage, an output device, and an input device. The device storage stores therein an image captured by each camera of the plurality of cameras. The output device displays an image exhibiting a positional relationship of the plurality of cameras. The input device sets tracking information for tracking a specific tracking target in the captured images to one camera among the plurality of cameras. When an imaginary movement line is input on the image exhibiting the positional relationship of the cameras, the input device sets the tracking information to one or more cameras corresponding to the imaginary movement line among the plurality of cameras other than the one camera.

Abstract: A stereoscopic control method includes: establishing a specific mapping relation between a specific disparity value and a specific set of a first focal setting value of a first sensor of a stereo camera and a second focal setting value of a second sensor of the stereo camera; and controlling stereoscopic focus of the stereo camera according to the specific mapping relation. Besides, a stereoscopic control apparatus includes a mapping unit and a focus control unit. The mapping unit is arranged for establishing at least a specific mapping relation between a specific disparity value and a specific set of a first focal setting value of a first sensor of a stereo camera and a second focal setting value of a second sensor of the stereo camera. The focus control unit is arranged for controlling stereoscopic focus of the stereo camera according to the specific mapping relation.

Abstract: Implementations are provided that relate, for example, to view tiling in video encoding and decoding. A particular method includes accessing a video picture that includes multiple pictures combined into a single picture (826), accessing information indicating how the multiple pictures in the accessed video picture are combined (806, 808, 822), decoding the video picture to provide a decoded representation of at least one of the multiple pictures (824, 826), and providing the accessed information and the decoded video picture as output (824, 826). Some other implementations format or process the information that indicates how multiple pictures included in a single video picture are combined into the single video picture, and format or process an encoded representation of the combined multiple pictures.

Abstract: An agricultural material baling system comprises, in one example, a bale forming component configured to form a bale of agricultural material from a terrain, and a control system configured to determine that the bale is to be released from the baling system onto the terrain, determine that a current location of the baling system has a slope above a threshold, determine a different location, that is spaced apart from the current location, for releasing the bale onto the terrain, and provide an output indicative of the different location. In one example, the control system is configured to receive yield data indicative of a volume of agricultural material in a path of the baler and to control the baling system based on the yield data. In one example, the yield data is obtained from a raking operation that rakes the agricultural material into a windrow.

Abstract: A method synchronizes autofocus in a system having a master camera and a slave camera. The method comprises: focusing the slave camera based on a map and a result of an autofocus operation by the master camera, while capturing each of a plurality of images. The map relates a plurality of master camera lens positions of the master camera to corresponding slave camera lens positions of the slave camera. An autofocus operation is periodically performed in the slave camera to determine an additional slave camera lens position for an additional image. The map is adaptively updated, based at least partially on the additional slave camera lens position.

Abstract: A method and apparatus for the inspection of a hostile environment includes a sensorized device carrying a plurality of image sensors positioned with different orientations, so as to detect image data of the hostile environment; a support adapted to support the sensorized device in the hostile environment; a processor of the image data generating a spherical and/or three-dimensional image based on the image data; and a remote display device adapted to be positioned outside the hostile environment and in communication with at least the processor of the image data, the plurality of sensors detecting contemporaneous image data of at least 60% of 4? steradians of the hostile environment, the apparatus being adapted for a hostile environment with temperatures and/or atmospheric contaminants harmful or dangerous for human beings.

Abstract: An image processing device includes: a generating unit that generates, based on a distance image, two-dimensional distribution information indicating a two-dimensional distribution of an object, in which a distance in a horizontal direction, a distance in a depth direction, and a frequency value corresponding to the distances are related; and a labeling unit that conducts a labeling process by conducting search on the two-dimensional distribution information multiple times, detecting a pixel having the frequency value that is more than a predetermined threshold during each search, and assigning a label that is different in each times of search, wherein the labeling unit conducts a first labeling process by selecting a pixel having the frequency value that is more than the threshold from pixels that abut a pixel being searched in a search direction and assigning the label when the label is not assigned to the pixel selected.

Abstract: An unobstructed image portion of a captured image from a first camera of a camera pair, e.g., a stereoscopic camera pair including fisheye lenses, is combined with a scaled extracted image portion generated from a captured image from a second camera in the camera pair. An unobstructed image portion of a captured image from the second camera of the camera pair is combined with a scaled extracted image portion generated from a captured image from the first camera in the camera pair. As part of the combining obstructed image portions which were obstructed by part of the adjacent camera are replaced in some embodiments. In some embodiments, the obstructions are due to adjacent fisheye lens. In various embodiments fish eye lenses which have been cut to be flat on one side are used for the left and right cameras with the spacing between the optical axis approximating the spacing between the optical axis of a human person's eyes.

Abstract: Systems and methods for synchronizing frame captures for cameras with different fields of capture are described. An example device includes a first camera and second camera. The first camera includes a first camera sensor and a first rolling shutter. The first camera is configured to prevent scanning pixels from a first row to a row n of the first camera sensor and configured to begin sequentially scanning pixels of the row n of the first camera sensor. The second camera includes a second camera sensor and a second rolling shutter. The second camera is configured to begin sequentially scanning pixels of a first row of the second camera sensor concurrently with beginning to sequentially scan pixels of the row n of the first camera sensor. The first row of the second camera sensor corresponds to a row within a predefined number of rows after the first camera sensor's row n.

Abstract: Using the same image sensor to capture both a two-dimensional (2D) image of a three-dimensional (3D) object and 3D depth measurements for the object. A laser point-scans the surface of the object with light spots, which are detected by a pixel array in the image sensor to generate the 3D depth profile of the object using triangulation. Each row of pixels in the pixel array forms an epipolar line of the corresponding laser scan line. Timestamping provides a correspondence between the pixel location of a captured light spot and the respective scan angle of the laser to remove any ambiguity in triangulation. An Analog-to-Digital Converter (ADC) in the image sensor generates a multi-bit output in the 2D mode and a binary output in the 3D mode to generate timestamps. Strong ambient light is rejected by switching the image sensor to a 3D logarithmic mode from a 3D linear mode.

Abstract: An exemplary data precision preservation system accesses a depth representation of a virtual reality scene associated with a world coordinate space, divides the representation into sections associated with different clip coordinate spaces, and determines a lowest and a highest non-null depth value represented in a particular section. Based on these values, the system determines an inverse view-projection transform for transforming depth values from a clip coordinate space of the particular section to the world coordinate space. The system converts original depth values represented in the particular section to compressed depth values that are normalized so as to be represented using fewer data bits the original depth values. The system provides, to a media player device by way of a network, a virtual reality dataset representative of the virtual reality scene and including the compressed depth values and the inverse view-projection transform. Corresponding systems and methods are also disclosed.

Abstract: Using the same image sensor to capture both a two-dimensional (2D) image of a three-dimensional (3D) object and 3D depth measurements for the object. A laser point-scans the surface of the object with light spots, which are detected by a pixel array in the image sensor to generate the 3D depth profile of the object using triangulation. Each row of pixels in the pixel array forms an epipolar line of the corresponding laser scan line. Timestamping provides a correspondence between the pixel location of a captured light spot and the respective scan angle of the laser to remove any ambiguity in triangulation. An Analog-to-Digital Converter (ADC) in the image sensor generates a multi-bit output in the 2D mode and a binary output in the 3D mode to generate timestamps. Strong ambient light is rejected by switching the image sensor to a 3D logarithmic mode from a 3D linear mode.

Abstract: According to an embodiment, an actuation system includes a movable member, an actuator, at least one image taking unit, a coordinate conversion unit and a first actuation control unit. The actuator moves the movable member. The coordinate conversion unit performs coordinate conversion from an image-taking coordinate system of a taken image taken by the image taking unit to an arithmetic coordinate system for performing comparison. The first actuation control unit controls the actuator to reduce a deviation according to visual servo in the arithmetic coordinate system.

Abstract: The electronic apparatus includes: a plurality of image sensors including a first image sensor and a second image sensor; and a processor electrically connected to the plurality of image sensors and configured to output a read control signal and a synchronization signal to the plurality of image sensors, wherein the processor is further configured to: output a first read control signal to the first image sensor and receive first data read from the first image sensor; output a second read control signal to the second image sensor and store second data read from the second image sensor in a temporary memory; and output the second data stored in the temporary memory based on an output control signal generated between the first read control signal and a next first read control signal and generate merged data in which the first data and the second data are merged.

Abstract: An imaging system for producing images to be displayed to user via a head-mounted display apparatus that includes means for detecting gaze direction of user. The imaging system includes a first outer camera and a second outer camera, at least one inner camera, and processor coupled to aforesaid cameras and means for detecting gaze direction. The processor is configured to (i) obtain inter-pupillary distance of user with respect to user's gaze at infinity; (ii) receive detected gaze direction; (iii) control first outer camera, second outer camera and at least one inner camera to capture first outer image, second outer image and at least one inner image of a scene; and (iv) process first outer image and inner image to generate first view of the scene, and process second outer image and inner image to generate second view of the scene, based upon inter-pupillary distance and detected gaze direction.

Abstract: A method and apparatus for measuring a distance to an object, be a device (100) having at least two cameras (104, 106), is described. One or more first images including the object are acquired by a first camera of the device and one or more first reference images are acquired by a second camera of the device, while the device is in a first position. One or more second images including the object and one or more second reference images are acquired by cameras of the device, while the device is in a second position, different from the first position. Based on the first and second reference images, information on the displacement of at least one camera of the device between the first and second position are determined. The distance from the device to the object is calculated based on the first and second images including the object and the determined information on the displacement of the at least one camera.

Abstract: An infrared sensor including a plurality of infrared cameras is described. Each infrared camera includes an infrared detector and imaging optics. The imaging optics are arranged to image infrared light onto the infrared detector, the imaging optics having a field of view with a center optical axis, and field of view boundary lines bounding the field of view on sides of the field of view. The infrared cameras are arranged such that the center optical axes of the field of views intersect each other, and the fields of view overlap each other.

Abstract: A system or method for training may display a student avatar and an expert avatar. A method may include capturing movement of a user attempting a technique, and generating a student avatar animation from the captured movement. The method may include retrieving a 3D expert avatar animation corresponding to the technique. The method may include displaying the 3D student avatar animation and the 3D expert avatar animation. For example, the animations may be displayed concurrently.

Abstract: An in-vehicle camera apparatus is provided for use in a vehicle. The in-vehicle camera apparatus includes at least one image capturing unit configured to capture an image of an external environment of the vehicle, at least one housing configured to have the at least one image capturing unit mounted thereto, and a mounting part via which the at least one housing is to be mounted to the vehicle. The mounting part is made of a material having a lower coefficient of linear expansion than the at least one housing. The mounting part has at least one first portion configured to be fixed to the vehicle and a plurality of second portions configured to be fixed to the at least one housing.

Abstract: The present disclosure relates to methods and systems that facilitate determination of a pose of a vehicle based on various combinations of map data and sensor data received from light detection and ranging (LIDAR) devices and/or camera devices. An example method includes receiving point cloud data from a (LIDAR) device and transforming the point cloud data to provide a top-down image. The method also includes comparing the top-down image to a reference image and determining, based on the comparison, a yaw error. An alternative method includes receiving camera image data from a camera and transforming the camera image data to provide a top-down image. The method also includes comparing the top-down image to a reference image and determining, based on the comparison, a yaw error.

Abstract: The present invention provides a stereo camera in which it is possible to improve the precision of correcting vertical offset between a first image captured by a first image-capture unit and a second image captured by a second image-capture unit. In the present invention, an image-capture system unit 100a captures a benchmark image. An image-capture system unit 100b captures a reference image. A geometry correction unit 125 generates a plurality of geometry-corrected reference images having differing amounts of vertical-direction movement from the reference image. A parallax calculation unit 126 generates a plurality of parallax images from the combination of the benchmark image and each of the reference images. A parallax image evaluation unit 131 calculates evaluation values pertaining to the reliability of each of the parallax images.