Abstract: An image processing device is provided that can easily generate label data used for an object recognition technology using three-dimensional point cloud data. The image processing device (1) includes a three-dimensional point cloud obtaining unit (2), a two-dimensional label obtaining unit (4), and a label conversion unit (6). The three-dimensional point cloud obtaining unit (2) obtains three-dimensional point cloud data that represents a three-dimensional point cloud of an object. The two-dimensional label obtaining unit (4) obtains two-dimensional label data corresponding to a two-dimensional image of the object. The label conversion unit (6) associates the two-dimensional label data with the three-dimensional point cloud data, and converts the two-dimensional label data into three-dimensional point cloud label data that indicates a label of the three-dimensional point cloud.

Abstract: A method for using a trained generative adversarial network to improve vehicle orientation and navigation includes loading a semantically-segmented 3D point cloud into a virtual reality simulation environment; processing the 3D point cloud; and displaying an output including at least one attribute. A computing system for using a trained generative adversarial network to improve vehicle orientation and navigation includes one or more processors, and one or more memories having stored thereon computer-executable instructions that, when executed, cause the computing system to: load a semantically-segmented 3D point cloud into a virtual reality simulation environment; process the 3D point cloud; and display an output including at least one attribute.

Abstract: A detection device (10) includes a transmitter (110), a receiver (120), and a control unit (140). The transmitter (110) is capable of transmitting an electromagnetic wave. The receiver (120) is capable of receiving a signal (S1), that is, the electromagnetic wave transmitted from the transmitter (110) and reflected by an object (O). The control unit (140) controls the size of an instantaneous field of view (so-called IFOV) of the receiver (120).

Abstract: The present disclosure relates to an image processing apparatus, an image processing method, and a program for performing lens emulation more easily. An aberration table is generated as a table including converging ray vectors derived from aberrations of a lens selected through a user interface and targeted for emulation. A light focusing process is performed to generate a virtual captured image taken of an object through the selected lens, by use of the aberration table and a multi-view image. The present disclosure may be applied, for example, to image processing apparatuses, information processing apparatuses, imaging apparatuses, electronic devices, information processing methods, or programs.

Abstract: Stereoscopy in which at least one image of a scene is recorded from a first viewing angle, and at least one image of the scene is recorded from a second viewing angle. The scene is recorded multiple times from the first viewing angle. A first combination image is obtained from the various images recorded from the first viewing angle, said combination image according to the stipulation of a comparison algorithm having smaller differences in relation to at least one image also recorded from the second viewing angle or smaller differences in relation to a second combination image obtained from the images from the second viewing angle than each individual image recorded from the first viewing angle. An image of the scene with the depth information is obtained from the first combination image and at least one image from the second viewing angle or the second combination image.

Abstract: A method for localizing, in a space containing at least one determined object, an object element associated to a particular 2D representation element in a determined 2D image of the space, may have: deriving a range or interval of candidate spatial positions for the imaged object element on the basis of predefined positional relationships); restricting the range or interval of candidate spatial positions to at least one restricted range or interval of admissible candidate spatial positions, wherein restricting includes at least one of: limiting the range or interval of candidate spatial positions using at least one inclusive volume surrounding at least one determined object; and limiting the range or interval of candidate spatial positions using at least one exclusive volume surrounding non-admissible candidate spatial positions; and retrieving, among the admissible candidate spatial positions of the restricted range or interval, a most appropriate candidate spatial position on the basis of similarity metrics

Abstract: An orthognathic system using a 3-dimensional hybrid image constructing program according to an embodiment of the present invention includes: a dental clinic server configured to prepare preliminary data for orthognathic surgery based on CT image data acquired by a CT imaging apparatus, transmit the preliminary data to an orthognathic device manufacturer server, perform virtual orthognathic surgery by inputting osteotomy data received from the orthognathic device manufacturer server into the 3-dimensional hybrid image constructing program, and transmit the result of the virtual orthognathic surgery to the orthognathic device manufacturer server; and an orthognathic device manufacturer server configured to merge CT data with gingival data, extract this to generate a jaw-bone 3-dimensional image file, cut the jaw bones within the program by using the generated jaw-bone 3-dimensional image file, and transmit the data to the dental clinic server.

Abstract: A slope monitoring system for monitoring deformation of a wall or slope from a virtual point of view. A slope monitoring device directs signals at a slope or wall and compiles direction data, range data and amplitude data from signals reflected from the slope or wall. The direction data, range data and amplitude data comes from a first real position and a second real position. The data may be recorded sequentially or concurrently.

Abstract: The invention relates to an optical measuring method for the three-dimensional detection of the surface of an object using an optical recording unit. The optical recording unit is moved relative to the object during a first measurement time interval, height maps are successively detected by the recording unit at a recording frequency, and at least some of the detected height maps are each added to an overall height map and displayed. The recording frequency is regulated by control signals during the measurement time interval. The control signals are generated spaced apart in time and for the purposes of producing each control signal, a statistic for the quality of the height map is determined for the respectively last detected height image and used for producing the control signal.

Abstract: An example computing system is configured to (i) present a three-dimensional (3D) visualization of a 3D model of a construction project; (ii) while presenting the 3D visualization of the 3D model of the construction project, present (a) a sectioning plane that defines a view of the 3D visualization of the 3D model of the construction project and (b) a sectioning control tool comprising an idealized 3D model, the sectioning control tool configured to set a location of the sectioning plane; (iii) receive user input indicating an interaction with the idealized 3D model; and (iv) based on the user input, adjust the location of the sectioning plane relative to the 3D visualization model of the 3D model of the construction project and thereby adjust the view of the 3D visualization of the 3D model of the construction project.

Abstract: A computer system displays a representation of a previously-captured media item, which includes or is associated with depth information corresponding to a physical environment in which the media item was captured. While displaying the representation of the media item, the system receives, via one or more input devices, one or more inputs corresponding to a request to display a representation of a measurement corresponding to a respective portion of the physical environment captured in the media item, and, in response, the system displays the representation of the measurement over at least a portion of the representation of the media item that corresponds to the respective portion of the physical environment, based on the depth information, and a label corresponding to the representation of the measurement that describes the measurement based on the depth information.

Abstract: A system for monitoring and recording and processing an activity includes one or more cameras for automatically recording video of the activity. A processor and memory associated and in communication with the camera is disposed near the location of the activity. The system may include AI logic configured to identify a user recorded within a video frame captured by the camera. The system may also detect and identify a user when the user is located within a predetermined area. The system may include a video processing engine configured to process images within the video frame to identify the user and may modify and format the video upon identifying the user and the activity. The system may include a communication module to communicate formatted video to a remote video processing system, which may further process the video and enable access to a mobile app of the user.

Abstract: Provided is a size measurement system for measuring the size of a user's foot, said system characterized by comprising: a size measurement instrument which is formed from a stand for the user to place the foot thereon, and a plurality of markers which are positioned in the periphery of the stand; and a measurement terminal which captures an image of a state in which the user has placed the foot on the stand in the size measurement instrument such that the plurality of markers can be recognized from a plurality of directions, and which computes size data for the foot from contour information for the foot obtained through the image capture from each of the directions

Abstract: Provided is a real-time omnidirectional stereo matching method in a camera system including a first pair of fisheye cameras including first and second fisheye cameras provided to perform shooting in opposite directions and a second pair of fisheye cameras including third and fourth fisheye cameras provided to perform shooting in opposite directions and in which the first pair of fisheye cameras and the second pair of fisheye cameras are vertically provided, including receiving fisheye images of a subject captured through the first to the fourth fisheye cameras; selecting one fisheye camera from among fisheye cameras for each pixel of a preset reference fisheye image among the fisheye images using a sweep volume for preset distance candidates; generating a distance map for all pixels using the reference fisheye image and a fisheye image of the one fisheye camera; and performing real-time stereo matching on the fisheye images using the distance map.

Abstract: Systems with a contouring method are provided for contouring one or more targets that correspond to specific organs and/or tumors in a three-dimensional medical image of a patient using neural networks. The contouring system includes a storage unit, a processing unit, and a plurality of modules that are computer operable. The processing unit is used to obtain the image, and then to generate one or more contouring images using a contouring method. The contouring method includes enhancing image features and improving contouring accuracy using an image preprocessing module, and extracting a plurality of multi-scale image representations and expanding these representations to one or more contouring images using a neural network-based contouring module.

Abstract: A method for accelerating hyperspectral video reconstruction includes steps of: acquiring, according to a spectral video and an RGB video captured by a hyperspectral video camera, a calibration matrix of the spectral video and the RGB video; sorting the calibration matrix to generate an ordered calibration matrix; converting, according to the ordered calibration matrix, the spectral video and the RGB video into a data matrix in a parallel manner; acquiring all related calibration points of a reconstruction region according to the ordered calibration matrix; and, reconstructing a hyperspectral video in a parallel manner according to the related calibration points and the data matrix. The related calibration points are acquired by sorting the calibration matrix, such that the number of times the calibration matrix is traverse is reduced, and the computation amount of hyperspectral video reconstruction is decreased.

Abstract: A topographical measurement system uses an imaging cartridge formed of a rigid optical element and a clear, elastomeric sensing surface configured to capture high-resolution topographical data from a measurement surface. The imaging cartridge may be configured as a removable cartridge for the system so that the imaging cartridge, including the rigid optical element and elastomeric sensing surface can be removed and replaced as a single, integral component that is robust/stable over multiple uses, and easily user-replaceable as frequently as necessary or desired. The cartridge may also usefully incorporate a number of light shaping and other features to support optimal illumination and image capture.

Abstract: A high-accuracy three-dimensional reconstruction method and system, a computer device, and a storage medium are provided. The method includes: performing calibration on an imaging apparatus; calculating each unidirectional absolute phase distribution diagram of a planar target; establishing an imaging apparatus coordinate system, and fitting corresponding epipolar lines in a normalization plane; calculating intersections between the corresponding planar target and rays formed by points on an epipolar line and an optical center of the imaging apparatus, fitting a projection beam, and establishing a projection mapping coefficient table; and projecting a pattern to an object under test, acquiring an object image of the object under test by using the imaging apparatus, calculating and searching for projection mapping coefficients corresponding to absolute phases in the object image, and calculating corresponding spatial three-dimensional point coordinates by using the projection mapping coefficients.

Abstract: A system for creating a context for a virtual object is disclosed, comprising a processing unit, wherein the processing unit is configured to receive values of the one or more parameters from the input unit, create a three-dimensional textual matrix of sensor representation of a real environment world based on the one or more parameters, in real-time, wherein the processing unit create textual information that is reusable in the three-dimensional matrix, and create a context of a specific virtual object from one or more virtual objects with respect to the real-world environment based on the three-dimensional textual matrix, wherein the context of the specific virtual object is created based on the context of one or more other virtual objects.

Abstract: The present invention discloses a disparity estimation optimization method based on upsampling and exact rematching, which conducts exact rematching within a small range in an optimized network, improves previous upsampling methods such as neighbor interpolation and bilinear interpolation for disparity maps or cost maps, and works out a propagation-based upsampling method by the way of network so that accurate disparity values can be better restored from disparity maps in the upsampling process.

Abstract: The present disclosure relates to a method for determining a position and an orientation.

Abstract: Techniques for machine learning based image attribute determination are disclosed. In some embodiments, one or more unknown attributes of a received image comprising a prescribed environment are determined using a machine learning based framework. The machine learning based framework is at least in part trained on image data sets comprising a model environment that substantially simulates the prescribed environment.

Abstract: Described are systems and methods for detecting objects using calibrated imaging devices and obfuscating, in real-time or near real time, portions of the video data to protect the privacy of operators represented in the video data. For example, a position of an operator within a fulfillment center may be determined or tracked in video data and the pixels representative of that operator may be obfuscated using pixilation and/or other techniques so that a reviewing agent that is viewing the video data cannot determine the identity of the operator. Such obfuscation may be performed in real-time or near real-time using automated processing. In addition, only portions of the video data may be obfuscated so that events (e.g., item picks, item place) and/or other objects represented in the video data are still viewable to the reviewing agent.

Abstract: A horizontal cone shift model includes a baseboard having two mounting openings in parallel orientation to a width of the baseboard. A background board is attached to the baseboard along its length to represent a radiographic film. A plurality of magnetic labels are provided for placement on the baseboard and background board. A first rod and a second rod fit into each of the two mounting openings. The first rod is distinguishable from the second rod. A light source is located on the baseboard and is movable along the length of the baseboard such that shadows of the first rod and second rod appear on the background board. The plurality of labels are used to identify the first and second rods and locations.

Abstract: A system and method for performing real-time quality inspection of objects is disclosed. The system and method include a transport to move objects being inspected, allowing the inspection to be performed in-line. At least one optical acquisition unit is provided that captured optical images of the objects being inspected. The captured optical images are matched to CAD models of objects, and the matched CAD model is extracted. A laser with an illumination light beam has a wavelength in the violet or ultraviolet range then conducts scans of the objects, which are formed into three-dimensional point clouds. The point clouds are compared to the extracted CAD models for each object, where CTF are compared to user input or CAD model information and the object is determined to be acceptable or defective based on the extent of deviation between the point cloud and the CAD model.

Abstract: A method and a device for performing plane detection are provided. The method includes acquiring an input image, extracting features of the input image using a deep neural network and estimating a depth map of the input image based on the extracted features, performing region segmentation using the depth map to detect plane regions in the input image.

Abstract: Provided is a method for processing images. The method includes: acquiring a plurality of original sub-images acquired by a plurality of cameras in a camera array; converting first pixel coordinates of each pixel in each original sub-image in an image coordinate system into target pixel coordinates in a target plane coordinate system; determining a region pixel value of each sub-region according to a correspondence between the target pixel coordinates of each pixel in each original sub-image and each sub-region in the target plane; and determining a spliced image in the target plane based on the region pixel value of each sub-region.

Abstract: Image processing to discriminate imaged objects that are adjacent or overlapping. Non-empty cells of the image that contain portions of the objects, and empty cells that lack any portions of the objects, are all determined. A global convex hull is defined to surround the non-empty cells of the image. Voids, including at least a first void and a second void, are found within the global convex hull, each being composed of contiguous empty cells and having a corresponding void boundary. A separation line is defined based on a first separation line endpoint along the void boundary of the first void and a second separation line endpoint along the void boundary of the second void, to separate two of the objects in the image. An output may be produced that includes indicia of at least portions of distinct boundaries of the objects in the image based on the separation line.

Abstract: Implementations are described herein for performing sparse depth estimation in the agricultural domain. In various implementations, one or more spatial aspects of a first visual annotation of a first candidate plant trait instance depicted in a first stereo image may be compared to one or more spatial aspects of a second visual annotation of a second candidate plant trait instance depicted in a second stereo image. Based on the comparison, it may be determined that the first and second candidate plant trait instances are a match for the same plant trait instance across the first and second stereo images. A disparity may be determined between the first and second candidate plant trait instances. Based on the disparity, a depth associated with the same plant trait instance may be estimated and used for various purposes.

Abstract: A driving assistance method for a vehicle. An instantaneous speed of the vehicle and an instantaneous yaw rate of the vehicle are ascertained. An operation of self-locating of the vehicle is carried out on the basis of the ascertained, instantaneous speed and the ascertained, instantaneous yaw rate of the vehicle. To that end, an instantaneous circumferential wheel speed of one or more wheels of the vehicle is directly measured, evaluated and taken as a basis of the determination of the instantaneous speed and the instantaneous yaw rate of the vehicle.

Abstract: A distance measurement device includes: an image capturer that captures N segmental images corresponding to N segmental distances into which a distance measurement range is divided; and a range image generator that generates a range image from the N segmental images. The range image generator determines: among segmental pixels included in the N segmental images, a segmental pixel having a maximum signal value from N segmental pixels at the same pixel position among pixel positions; a value indicating a segmental distance of the segmental pixel having the maximum signal value to be a distance value of the pixel position of the range image, when the maximum signal value is greater than or equal to a threshold; and a value indicating a value outside the distance measurement range to be the distance value of the pixel position of the range image, when the maximum signal value is less than the threshold.

Abstract: Techniques for generating an image are disclosed. In some embodiments, a received input image is transformed to generate an output image using a machine learning based framework that is trained on a constrained set of images. The generated output image comprises an attribute learned by the machine learning based framework from the set of images.

Abstract: A machine learning based image processing and generation framework is disclosed. In some embodiments, depth values of an object or asset in a received input image are at least in part determined using a machine learning based framework that is constrained to a known prescribed environment. Determined depth values facilitate generation of other views of the object or asset.

Abstract: A medical image processing method includes: obtaining a biological tissue image including a biological tissue, recognizing, in the biological tissue image, a first region of a lesion object in the biological tissue; recognizing a lesion attribute matching the lesion object; dividing an image region of the biological tissue in the biological tissue image into a plurality of quadrant regions; obtaining target quadrant position information of a quadrant region in which the first region is located; and generating medical service data according to the target quadrant position information and the lesion attribute.

Abstract: An information processing method includes: three-dimensional (3D) point information of a 3D point cloud is obtained; a two-dimensional (2D) point cloud image from projection of the 3D point cloud on a horizontal plane is generated based on the 3D point information; and projection coordinates of 3D points comprised in the 3D point cloud in a reference coordinate system of a reference plane graph are determined based on a consistency degree that the 2D point cloud image has with the reference plane graph, where the reference plane graph is used for representing a projection graph with reference coordinates that is obtained through the projection of a target object on the horizontal plane, and the 3D point cloud is used for representing 3D space information of the target object.

Abstract: A manufacturing monitoring assistance device includes: a model creation unit creating a computation model when a product as a sample is normal, based on a three-dimensional form acquired from the product; a simulation unit creating a corrective computation model when the product is abnormal, by adding a sample of an abnormal portion in the product to the created computation model, and performing a simulation on each of the computation model and the corrective computation model; and a monitoring method determination unit determining a method for monitoring a manufacturing process for the product, based on an abnormality index being a difference between an output from a sensor as a result of the simulation performed on the computation model and an output from a sensor as a result of the simulation performed on the corrective computation model, and causing an output device to display the determined method and the abnormality index.

Abstract: An exemplary method includes a scene annotation system accessing image scan information of a scene captured by an imaging device of a computing device, generating, based on the image scan information, depth data associated with one or more objects in the scene, providing a two-dimensional (2D) annotation grid for concurrent display together with an image of the scene, at least part of the 2D annotation grid mapped to the depth data, receiving a user input (e.g., via hand gesture, touch screen, mouse, and/or keyboard) that selects a portion of the 2D annotation grid and a corresponding position within the scene where a digital object will be anchored based on the depth data, and instructing the computing device to display, by way of a display device of the computing device and based on the user input, the digital object anchored at the corresponding position within the scene.

Abstract: The invention relates to a computer-implemented method for assessing a clearance between a tooth element to be provided with a dental restoration element and an antagonist of the tooth element. A 3D digital teeth model with 3D digital representations of the tooth element and of a surface area of the antagonist is received. The surface area is offset from the antagonist by adding an offset. The resulting offset surface area defines a limit for the tooth element. For the clearance it is checked, whether the tooth element violates one or more penetration criteria regarding the offset surface area. If the tooth element violates the one or more penetration criteria, the clearance is assessed to be insufficient for the tooth element to be provided with the dental restoration element. Else the clearance is assessed to be sufficient. The result of the assessment is output.

Abstract: A medical image processing device a reference image that is a medical image with which boundary line information related to a boundary line that is a boundary between an abnormal region and a normal region and landmark information related to a landmark that is a characteristic structure of the subject are associated and a captured image that is the medical image captured in real time, detects the landmark from the captured image, calculates a ratio of match between the landmark included in the reference image and the landmark included in the captured image, estimates a correspondence relationship between the reference image and the captured image on the basis of the ratio of match and information regarding the landmarks included in the reference image and the captured image, and generates a superimposition image in which the boundary line associated with the reference image is superimposed on the captured image on the basis of the correspondence relationship.

Abstract: Methods and apparatus to monitor environments are disclosed. Example audience measurement devices disclosed herein execute, in connection with a first frame of data, a three-dimensional recognition analysis on an object detected in an environment within a threshold distance from a sensor. Disclosed example audience measurement devices also detect that the object has moved outside the threshold distance from the sensor in a second frame subsequent to the first frame. Disclosed example audience measurement devices further execute a two-dimensional recognition analysis on the object in the second frame.

Abstract: An apparatus and method for data fusion between heterogeneous sensors are disclosed. The method for data fusion between the heterogeneous sensors may include identifying image data and point cloud data for a search area by each of a camera sensor and a LiDAR sensor that are calibrated using a marker board having a hole; recognizing a translation vector determined through calibrating of the camera sensor and the LiDAR sensor; and projecting the point cloud data of the LiDAR sensor onto the image data of the camera sensor using the recognized translation vector to fuse the identified image data and point cloud data.

Abstract: According to one embodiment, a system for determining a position of a vehicle includes an image sensor, a top-down view component, a comparison component, and a location component. The image sensor obtains an image of an environment near a vehicle. The top-down view component is configured to generate a top-down view of a ground surface based on the image of the environment. The comparison component is configured to compare the top-down image with a map, the map comprising a top-down light LIDAR intensity map or a vector-based semantic map. The location component is configured to determine a location of the vehicle on the map based on the comparison.

Abstract: A depth estimation from focus method and system includes receiving input image data containing focus information, generating an intermediate attention map by an AI model, normalizing the intermediate attention map into a depth attention map via a normalization function, and deriving expected depth values for the input image data containing focus information from the depth attention map. The AI model for depth estimation can be trained unsupervisedly without ground truth depth maps. The AI model of some embodiments is a shared network estimating a depth map and reconstructing an AiF image from a set of images with different focus positions.

Abstract: A method and system for segmenting a plurality of images. The method comprises the steps of segmenting the image through a novel clustering technique that is, generating a composite depth map including temporally stable segments of the image as well as segments in subsequent images that have changed. These changes may be determined by determining one or more differences between the temporally stable depth map and segments included in one or more subsequent frames. Thereafter, the portions of the one or more subsequent frames that include segments including changes from their corresponding segments in the temporally stable depth map are processed and are combined with the segments from the temporally stable depth map to compute their associated disparities in one or more subsequent frames. The images may include a pair of stereo images acquired through a stereo camera system at a substantially similar time.

Abstract: Proposed is a multi-view camera-based iterative calibration method for generation of a 3D volumetric model that performs calibration between cameras adjacent in a vertical direction for a plurality of frames, performs calibration while rotating with the results of viewpoints adjacent in the horizontal direction, and creates a virtual viewpoint between each camera pair to repeat calibration. Thus, images of various viewpoints are obtained using a plurality of low-cost commercial color-depth (RGB-D) cameras. By acquiring and performing the calibration of these images at various viewpoints, it is possible to increase the accuracy of calibration, and through this, it is possible to generate a high-quality real-life graphics volumetric model.

Abstract: Subject matter regards generating a 3D point cloud and registering the 3D point cloud to the surface of the Earth (sometimes called “geo-locating”). A method can include capturing, by unmanned vehicles (UVs), image data representative of respective overlapping subsections of the object, registering the overlapping subsections to each other, and geo-locating the registered overlapping subsections.

Abstract: An apparatus includes a generation unit configured to generate map information including a position of a feature point and identification information on an index in an image of a real space captured by a capturing apparatus, a collation unit configured to collate the identification information on the index in the generated map information with the identification information on the index in one or more pieces of registered map information, and to extract map information from the one or more pieces of registered map information based on a result of the collation, and an estimation unit configured to estimate a position and orientation of the capturing apparatus based on the position of the feature point in the extracted map information and the position of the feature point in the generated map information.

Abstract: A vehicular trailer guidance system includes a human machine interface provided in the vehicle and operable by a driver of the vehicle during a backing up maneuver of the vehicle with a trailer hitched thereto. The human machine interface comprises a rotary knob and operates (i) in a backing up trailer left-backup mode when rotated counter-clockwise, (ii) in a backing up trailer straight-backup mode when the rotary knob is pushed or pulled and (iii) in a backing up trailer right-backup mode when rotated clockwise. Responsive to selection by the driver of the backup mode, the system sets a desired trailer angle relative to the longitudinal axis of the vehicle to an angle that is commensurate with a driver-selected setting of the rotary knob. The system controls steering of the vehicle to back up the trailer to have the determined trailer angle coincide with the set desired trailer angle.

Abstract: The present invention provides a method for designing of a piece of apparel in particular an upper of a shoe, comprising the steps of providing at least one first panel including a plurality of feature points in an essentially two-dimensional configuration, arranging the at least one first panel on a first reference body in a three-dimensional configuration representing the piece of apparel to be designed, generating a first mapping between the two-dimensional configuration of the at least one first panel and the three-dimensional configuration of the at least one first panel using the plurality of feature points and designing the piece of apparel using the first mapping.

Abstract: Disclosed are systems, methods, and non-transitory computer-readable media for continuous surface and depth estimation. A continuous surface and depth estimation system determines the depth and surface normal of physical objects by using stereo vision limited within a predetermined window.