ADAPTIVE STITCHING OF FRAMES IN THE PROCESS OF CREATING A PANORAMIC FRAME

BACKGROUND

The present invention, in some embodiments thereof, relates to image processing and, more specifically, but not exclusively, to stitching of frames.

Virtual Reality (VR) is a special type of image or video content. VR, as its name suggests, is designed to replace the reality in order to provide the viewer with immersive sensation of the recorded content, including video and audio. The viewer uses a special type of display glasses, commonly referred to as VR headsets, VR goggles or VR glasses. The VR headset effectively blocks the viewer's natural vision and replaces it by a recorded or live-broadcasted content.

VR content is different than standard digital content designed to be presented on a flat screen, because VR is designed to replace the natural vision. VR is designed to be presented for a wide field of view (FOV) while providing stereo vision.

SUMMARY

According to some embodiments of the present invention, a method of creating a panoramic frame by combining frames along an irregular line identified in an overlapping depicted in the frames, comprising:

identifying an overlap between content documented in a first frame portion of a first frame and content documented in a second frame portion of a second frame;

segmenting a plurality of objects in the first frame portion;

identifying boundaries of the plurality of objects;

identifying an irregular line path connecting between two opposing sides of the first frame portion along at least some of the boundaries of at least some of the plurality of objects; and

stitching the first frame and second frame along the irregular line path in the process of creating a panoramic frame.

Optionally, the method comprises detecting in the first frame a plurality of feature points each having at least one predefined feature point characteristic; wherein the plurality of objects are segmented according to the plurality of feature points.

More optionally, the plurality of objects are identified on a watershed image generated from the first portion based on the plurality of feature points.

More optionally, the plurality of feature points are identified using an accelerated segment test (FAST) process.

Optionally, the overlap is identified by conducting a feature descriptor matching process.

Optionally, the first frame and the second frame are captured by imagers from a plurality of imagers of an imaging device.

More optionally, the first frame and the second frame are captured by two neighboring imagers from the plurality of imagers, a first field of view of one of the two neighboring imagers is overlapping with a second field of view of another of the two neighboring imagers; wherein the two neighboring imagers are mounted around a common center together with other imagers.

More optionally, an optical axis of a first of the two neighboring imagers is tilted in relation to an axis passing through the common center and a respective tangential point of an origin point of a field of view of the first neighboring imager on a virtual circle passing via all origin points of the fields of view of the plurality of imagers.

Optionally, the method further comprises: adjusting the irregular line path according to a new irregular line path calculated similarly to the irregular line path for a pair of new frames captured after the first frame and second frame, and using the adjusted irregular line path for stitching the new images.

Optionally, the method comprises the adjusting and the using are performed when a difference between the irregular line path and the new irregular line path is greater than a threshold.

Optionally, the panoramic frame is one of a plurality of frames of a virtual reality (VR) file.

More optionally, the panoramic frame is a left eye panoramic frame and the plurality of imagers are intertwined with a plurality of additional imagers capturing a plurality of additional frames which are combined into a right eye panoramic frame.

Optionally, the method further comprises combining the right eye panoramic frame with the left eye panoramic frame for creating a stereoscopic frame.

More optionally, members of the plurality of imagers and the plurality of additional frames are alternately arranged along a virtual circle encircling a common center.

More optionally, the panoramic frame is a right eye panoramic frame and the plurality of imagers are intertwined with a plurality of additional imagers capturing a plurality of additional frames which are combined into a left eye panoramic frame; further comprising combining the left eye panoramic frame with the right eye panoramic frame for creating a stereoscopic frame.

More optionally, the plurality of imagers are arranged along a virtual circle encircling a common center; wherein each one of the plurality of imagers is mounted such that an optical axis thereof is tilted in relation to an axis passing through the common center and a tangential point of an origin of a field of view of a respective the imager.

According to some embodiments of the present invention, a system of creating a panoramic frame by combining frames along an irregular line identified in an overlapping depicted in the frames. The system comprises an interface adapted for receiving first and second frames from two of a plurality of frames captured by a plurality of imagers of an imaging device, a code store adapted for storing a code, and a processor adapted for executing the code, wherein the code comprising: code instructions for identifying an overlap between content documented in a first frame portion of a first frame and content documented in a second frame portion of a second frame, code instructions for segmenting a plurality of objects in the first frame portion, code instructions for identifying boundaries of the plurality of objects, code instructions for identifying an irregular line path connecting between two opposing sides of the first frame portion along at least some of the boundaries of at least some of the plurality of objects, and code instructions for stitching the first frame and second frame along the irregular line path in the process of creating a panoramic frame.

Unless otherwise defined, all technical and/or scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention pertains. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of embodiments of the invention, exemplary methods and/or materials are described below. In case of conflict, the patent specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and are not intended to be necessarily limiting.

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

Some embodiments of the invention are herein described, by way of example only, with reference to the accompanying drawings. With specific reference now to the drawings in detail, it is stressed that the particulars shown are by way of example and for purposes of illustrative discussion of embodiments of the invention. In this regard, the description taken with the drawings makes apparent to those skilled in the art how embodiments of the invention may be practiced.

In the drawings:

FIG. 1 is a flowchart of creating a panoramic frame using an irregular stitch line identified in an overlapping portion of neighboring frames captured by a plurality of imagers having multiple viewing angles around a common center region, in accordance with some embodiments of the present invention;

FIG. 2 is a block diagram of components of a system for creating a panoramic frame, for instance by executing the method depicted in FIG. 1, in accordance with some embodiments of the present invention;

FIG. 3A is a schematic illustration of an exemplary arrangement of imagers capturing frames at multiple viewing angles that are stitched into the panoramic frame using the systems and/or methods described herein, in accordance with some embodiments of the present invention;

FIG. 3B is a schematic lateral illustration of a virtual reality (VR) imaging device having an arrangement of imagers for capturing frames at multiple viewing angles and for stitching the captured frames into a panoramic frame using methods described herein, in accordance with some embodiments of the present invention;

FIGS. 3C and 3D are schematic illustrations of an overlap between fields of view of imagers having, respectively, tilted optical axes and non-tilted optical axes in relation to a radius of a virtual circle passing via mounting points of the imagers, in accordance with some embodiments of the present invention;

FIG. 4 is an exemplary overlapping area marked with feature points (dots), boundaries of objects (green colored lines), and a shortest irregular path which crosses the exemplary overlapping area along some of the boundaries, in accordance with some embodiments of the present invention; and

FIG. 5 is an exemplary overlapping area marked with boundaries of objects depicted in the exemplary overlapping area, in accordance with some embodiments of the present invention; and

FIGS. 6A-6C are exemplary images depicting the process of acquiring frames captured at multiple viewing angles, adapting the frames, and the created panoramic frame, in accordance with some embodiments of the present invention.

DETAILED DESCRIPTION

The present invention, in some embodiments thereof, relates to image processing and, more specifically, but not exclusively, to stitching of frames.

An aspect of some embodiments of the present invention relates to systems and/or methods (e.g., code executed by a processor of a computing device) for stitching frames or images (for brevity, referred to herein interchangeably) captured at multiple viewing angles, optionally around a common center region, by multiple imagers (e.g. cameras, image sensors) arranged into a panoramic frame or panoramic frame that depicts the environment surrounding the common center region, for brevity also referred to herein a common center. The stitching is performed to reduce visibility of the stitching seams, and to create an improved panoramic frame viewing experience to the user, which more closely resembles the captured surrounding environment.

The stitching of frames is performed along an irregular line identified in an overlapping area between fields of view of the frames. The irregular line is optionally identified along boundaries of objects or segmented areas which are located in overlapping area.

The frames to be stitched by embodiments of the present invention may be pre-processed (i.e., before the stitching) for alignment and registration. Overlapping areas between two consequent frames captured by imagers having fields of view which image a common portion of an environment, referred to herein also as neighboring imagers, are identified. The overlapping areas may not necessarily be identified from an analysis of an overlapping area in the fields of view but rather estimated using external methods.

Optionally, the overlapping areas are estimated by projecting a visual representation of the frames onto a virtual sphere skeleton model. Alternatively or additionally, the overlapping areas are estimated based on a calibration model (e.g., a mathematical model) defined for one or both of the imagers that capture the overlapping frames. The calibration model may be defined based on principal point parameter(s), focal length parameter(s), and/or fisheye distortion parameter(s).

The systems and/or methods described herein provide a technical solution to the technical problem of how to reduce the visibility of seams in a panoramic frame that depicts the environment surrounding the common center when the panoramic frame is created by combining frames captured at multiple viewing angles. The stitched frames may be frames of a VR file. Such VR files may be used in virtual reality systems, for example, presented to a user within a VR headset for viewing VR videos. The frames acquired for stitching together are captured by different cameras, which have different perspectives and/or different characteristics such as focus, exposure, white balance. The lenses used by the imagers (e.g., wide and/or fish-eye lenses) may apply visual distortions, such as barrel, pincushion and vignette, which create additional challenges in stitching the frames. A sub technical problem may be stitching of the frames captured by the different cameras, which have different characteristics and changing orientation (e.g. do to minor movements of fixed or adjustable cameras), in a manner to decrease or eliminate visibility of the seams. Visible seams reduce the natural looking or real feeling that the VR video is able to provide. Reducing or eliminating stitching distortions (e.g., in left and right panoramic frames, each designed to be viewed by respective left and right eyes) and/or different artifacts improves the VR video, by removing or reducing inconsistent, non-horizontal parallax, which would otherwise cause the viewer discomfort and/or nausea.

The systems and/or methods described herein tie mathematical operations (e.g., estimation of overlapping areas of frames, calculation of motion gradient(s), and frame stitching) to the ability of a processor to process digital images, for example, by stitching frames acquired at multiple viewing angles around a common center into a panoramic frame based on an irregular line crossing an overlaying area of the frames.

It should be noted that there are several methods of image registration and alignment, defining the exact areas of where the content created by different camera modules need to be stitched. Once the registration is calculated, the second essential part is to define how exactly the overlapping areas of connecting frames will be cut and fused.

There are several methods that define the cutting line. The simplest one is a straight vertical line cutting both overlapping frames along a straight line, at a selected coordinate. Adding a gradient transition from one overlapping frame to another is also used for adapting a straight line. For example, when connecting two frames, left and right, the transparency of the overlapping area of the left frame may gradually increase from left to right transparence on the left frame and the exact opposite in the right frame. All the straight cutting methods have one significant disadvantage, human vision is especially sensitive to well defined geometric figures, such as straight lines, so that any well-defined stitching lines are prominent to the user.

The systems and/or methods described herein relate to processing frames acquired at multiple viewing angles by imagers mounted around a common center. New data is created in the form of a panoramic frame, by stitching together the acquired frames. The panoramic frame may be stored in a memory device, and optionally played back to a user, for example, displayed in VT headgear. The panoramic frame may be incorporated in a video that includes multiple consequent panoramic frames.

The systems and/or methods described herein improve performance of a computer, for example, by using less memory and/or improving computation time in producing an improved digital image.

Accordingly, the systems and/or methods described herein are necessarily rooted in computer technology to overcome an actual technical problem arising in digital image processing.

Before explaining at least one embodiment of the invention in detail, it is to be understood that the invention is not necessarily limited in its application to the details of construction and the arrangement of the components and/or methods set forth in the following description and/or illustrated in the drawings and/or the Examples. The invention is capable of other embodiments or of being practiced or carried out in various ways.

The present invention may be a system, a method, and/or a computer program product. The computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present invention.

The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.

Computer readable program instructions described herein can be downloaded to respective computing/processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and/or a wireless network. The network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. A network adapter card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing/processing device.

Computer readable program instructions for carrying out operations of the present invention may be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++ or the like, and conventional procedural programming languages, such as the “C” programming language or similar programming languages. The computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present invention.

Aspects of the present invention are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer readable program instructions.

These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function/act specified in the flowchart and/or block diagram block or blocks.

The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions/acts specified in the flowchart and/or block diagram block or blocks.

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts or carry out combinations of special purpose hardware and computer instructions.

Reference is now made to FIG. 1, which is a flowchart of a method of stitching frames captured by neighboring imagers at multiple viewing angles into a panoramic frame, in accordance with some embodiments of the present invention. The method identify an irregular line in overlapping areas of frames by detecting boundaries of objects or any other segmented areas in the overlapping areas and identifying the shortest irregular line which crosses (e.g. divides into two) the overlapping areas. The method is repeated for each pair of frames captured substantially simultaneously by imagers of a panoramic imaging device. Reference is also made to FIG. 2, which is a block diagram of components of a system 200 that allows a user to capture individual panoramic frames or a video of a sequence of panoramic frames of an environment surrounding (at least in part) a common center region, using multiple viewing imagers each pointed toward a different angle around the common center region. The user may record the video for playback in a virtual reality setting, such as using a VR headset. The method of claim 1 may be implemented by system 200 of FIG. 2.

System 200 includes a computing unit 202 housed with imagers 212, for example a custom designed unit, referred to herein as a VR imaging device (see for example FIG. 3B which is an exemplary housing of a VR imaging device), or separate from a housing which includes the imagers 212, for instance a personal computer, a server, a mobile device, a wearable computer, or other implementations. Computing unit 202 includes one or more processor(s) 204, and a program store 206 storing code instructions for execution by processor(s) 204.

Processor(s) 204 may be, for example, a processing unit(s) (CPU), one or more graphics processing unit(s) (GPUs), field programmable gate array(s) (FPGA), digital signal processor(s) (DSP), and application specific integrated circuit(s) (ASIC). Processor(s) may be part of a processing unit that includes multiple processors (homogenous or heterogeneous) arranged for parallel processing, as clusters and/or as one or more multi core processing units.

Program store 206 store code instructions implementable by processor(s) 204, for example, a random access memory (RAM), read-only memory (ROM), and/or a storage device, for example, non-volatile memory, magnetic media, semiconductor memory devices, hard drive, removable storage, and optical media (e.g., DVD, CD-ROM). The instructions are optionally to implement the method described in FIG. 1.

Computing unit 202 includes or is in communication with a data repository 208, for example, a memory, a storage unit, a hard drive, an optical disc, a remote storage server, and/or a cloud server (e.g., accessed via a network connection). Data repository 208 may store the raw acquired frames (e.g., in raw frame repository 208A), store the frames adapted as described by the systems and/or methods described herein (e.g., in adapted frame repository 208B), and/or the created panoramic frames (e.g., in panoramic frame repository 208C).

Computing unit 202 includes a data interface 210 (e.g., physical, virtual and/or a software interface) for receiving frames acquired from each of multiple imagers 212 (e.g., digital cameras). Imagers 212 (e.g. red, green, blue (RGB) imagers) are arranged to acquire a set of frames that depict an environment, optionally at least 180 degrees, optionally 360 (horizontally and/or vertically) around a common center region, for example see FIG. 3A which is described in greater details below.

Optionally, imagers 212 are paired to capture frame for the left or right eyes, which may be presented to the different eyes, for example, using a VT headset. In such an embodiment, imagers may be divided to two groups, left eye group and right eye group. Members of the left eye group are in even places and members of the right eye group are in uneven places (assuming that places are distributed in a sequential order in a circle surrounding the common center region) or vice versa. In such embodiments, frames captured by imagers of the left eye group are stitched to form a left eye panoramic frame and frames captured by imagers of the right eye group are stitched to form a right eye panoramic frame. The stitching of each one of the left eye panoramic frame and the right eye panoramic frame is done separately, for instance as described below (overlapping areas are between frames captured by group members). The left eye panoramic frame and the right eye panoramic frame are combined to create a stereoscopic panoramic frame. For example, in the arrangements 304306 of FIG. 3A, the dashed rectangles may be members of the left eye group and the non-dashed rectangles are members of the right eye group.

Alternatively or additionally, imagers 212 capture frames that are displayed to both eyes simultaneously, for example, projected within a 180 degree theater.

Each one of the imagers 212 may have a wide angle lens designed to capture a wide field of view, for example, a fish eye lens. Exemplary imagers 212 are cameras with ultra-wide and/or vertical angle lenses that capture about 120 degrees horizontally and about 175 degree vertically, or other values.

The number of imagers 212 is selected to cover the designed environment (twice in embodiments wherein a stereoscopic panoramic frame is created), and may be based on the field of view that may be captured by the lenses of the cameras, for example, 4 imagers, or 8 imagers, 10 imagers, 16 imagers and/or any intermediate or larger number of imagers.

Reference is now made to FIG. 3A, which depicts exemplary implementations for the arrangement of imagers 312 (e.g., corresponding to imagers 212 as described with reference to FIG. 2), in accordance with some embodiments of the present invention. Imagers 312 are mounted around a common center 302. Imagers 312 are arranged to acquire frames at multiple viewing angles. The frames are stitched into a panoramic frame (using the systems and/or methods described herein) that depict the environment surrounding common center 302.

Each one of arrangements 304 and 306 includes eight imagers 312 arranged to cover 360 degrees around common center 302. Implementation 304 depicts imagers 312 arranged in a square 308 (or rectangular) format, including two imagers 312 per side of square 308, which may be paired to capture frames for the left and right eyes. See also FIG. 3B which is a lateral view of an exemplary quadratic housing with truncated corners of a custom designed unit that includes 4 pairs of lateral imagers 1312 where each pair is located at another truncated corner of the exemplary quadratic housing. Arrangement 306 includes imagers 312 arranged in a circle 310 (or oval) format, spaced apart along the circumference of circle 310. Imagers 312 may capture frames for the left and right eyes. It is noted that other implementation shapes may be used.

According to some embodiments of the present invention, imagers are divided to pairs wherein each pair is designed to capture a stereoscopic frame. In such embodiments, overlapping areas in pairs of stereoscopic frames are identified and used for creating a panoramic frame is described below. For brevity, a stereoscopic frame is referred to herein as a frame and a pair of imagers designed to capture a stereoscopic frame is referred to herein as an imager. For example, a VR imaging device having an arrangement of pairs of imagers for capturing stereoscopic frames is used where the fields of view are as depicted in FIG. 3C. Optionally, optical axis each imager in a pair of imagers is tilted toward the other imager of the pair of imagers, for instance toward the optical axis thereof. Optionally, the tiling of the optical axis is in relation to an axis passing through the common center and through a respective tangential point of an origin point of the field of view of respective imager on a virtual circle passing via all origin points of the fields of view of all imagers, for example see the circle depicted in FIGS. 3C and 3D. Optionally, the tilting is between 20 and 30 degrees, for instance 22 degrees as depicted in FIG. 3C. The tilting of imagers reduces the overlapping areas between the fields of views. This can be seen when comparing the imagers in FIG. 3C which depicts an arrangement of imagers wherein each imager is tilted toward its paired imager with FIG. 3D which depicts another arrangement of imagers wherein the optical axis of each imager is aligned to continue a radius of a virtual circle passing via mounting points of all imagers, for instance the points of origin of the optical axes of the imagers. When comparing FIG. 3C with FIG. 3D one can see that the darker areas, which are intersections of the fields of view of imagers, are larger when the optical axes of the imagers are tilted as described above.

Implementations 304 and 306 may have a substantially disc shape, in which imagers 312 are arranged along a plane. It is noted that other implementation profiles may be used, for example, a sphere, or half-sphere.

Additional imagers 312 may be positioned to face up or down (not shown).

Computing unit 202 and imagers 312 may be housed within a casing based on implementations 304 and/or 306, for example, as a standalone portable unit, which may be used by consumers at home.

Referring now back to FIG. 2, computing unit 202 includes a communication interface 214 (e.g., physical, software, and/or virtual) to communicate with one or more external devices, to store and/or present the created panoramic frames (e.g., videos), for instance a Wi-Fi™ module or a Bluetooth™ module. Exemplary external devices include a personal display device 216 (e.g., a VR headset), a storage device 218 to store the videos for future playback, and a server 220 (e.g., web server, storage server, video server) that may be communicated with over a network 222. It is noted that the recorded videos may be publicly played, for example, projected onto a panoramic screen in a theater, in a room, or at home, for example, by a projector (not shown).

Computing unit 202 includes or is in communication with a user interface 224 (which may be integrated within a housing containing computing unit 202, implemented as software on a client terminal, and/or implemented as part of the display device displaying the panoramic frames), for example, a touchscreen, a keyboard, a mouse, and voice activated software using speakers and microphone. User interface 224 may access a code (e.g., stored on a client terminal and/or on computing unit 202) to customize the creation of the panoramic frames based on user inputs.

According to some embodiments of the present invention, the imagers are calibrated for calculating a camera calibration model for image alignment as well as reduction of parallax distortion. Optionally, the camera calibration model is calculated based on intrinsic parameters of each imager, for instance principal point parameter(s), focal length parameter(s), and/or fisheye distortion parameter(s) and optionally based on extrinsic parameter(s). For example, intrinsic parameters may include fisheye distortion parameters.

In use, the parameters may be calculated by placing imagers of the VR imaging device in front of a chessboard pattern while the VR imaging device is rotated and capturing sequence(s) of frames.

During calculation of a camera calibration model, which is optionally executed using processor(s) 204; corners of a pattern of n×m chessboard tiles are detected, for instance by finding a linear least squares homography of the pattern. A Gauss-Newton method may be applied to find the above imager parameters and rotations and translations of the imagers that yield the detected homographies for several views of chessboard. A Jacobian matrix is calculated and a quality criterion is calculated based on a mean square error method. This allows calibrating extrinsic parameters of a respective rig for the calculation of the camera calibration model, mutual rotations by the means of bundle adjustment on frames where distant objects are shot. For example, the extrinsic parameters are angles of rotation of each imager (e.g. optical axis angle) in 3D space (e.g. tilt, pan, and roll). Distant content may be detected by calculating an optical flow between each two imagers looking in the same direction by displaced horizontally (a stereo pair). Parallax of pairs of frames containing distant content should have a significantly lower parallax. Homographies may be detected by matching feature points, for instance using a scale-invariant feature transform (SIFT) process or a speeded up robust features (SURF) process. Rotations may be found by a Levenerg-Macart method. Jacobian matrixes are numerically approximated. Intrinsic parameters may not be changed at this stage. The calibration may be initially executed using twenty frames of the chessboard pattern per imager. Optionally, unclear frames with high pixel re-projection error of calibration corners are removed to assure only frames having a quality above a threshold are used for the calibration in order to ensure a low pixel re-projection error. As indicated above, the calibration is made on the corners of a checker board pattern item. It is optionally assumed that the board is not moving, and hence remains in fixed coordinates in world coordinate system (X-Y plane) with squares starting at (0,0,0). A bundle adjustment algorithm may be applied.

Optionally, the following pseudo code iteratively collects frames for calculating intrinsic parameters:

For each frame: when a chessboard is detected in the frame:

[RMS, K, DistortionCoeffs, rotation_vectors_of_cameras, translation_vectors_of_cameras]

=calibrate(board squares world-coordinates, board squares image coordinates in ALL image that were collected so far, fix the skew to 0)

if (calibration succeeded) AND (RMS<3 pix OR this is a first frame to image a chessboard) than update images collection with the current image

Finally, return the solution that comes out from all frames captured so far.

Optionally, the following pseudo code is used for calculating extrinsic parameters:

BEGIN

For each original distorted image

1. Find distorted features.

2. For each feature (e.g. corners of the chess board) undistorted location is calculated using the distortion coefficients (from the intrinsic parameters).

3. Match all features (with undistorted feature point locations).

4. Solve extrinsic orientation using bundle adjustment on the match received from step 3.

5. Calculate root mean square (RMS) of a received solution to step 4 (optionally, a frame is added only when the RMS is above a threshold).

6. Rotate imagers so that a desired ‘forward camera’ rotation matrix is I_3×3.

END

Optionally, 2-5 are executed using standard computer vision routines, using the intrinsic parameters of the respective imager calculated during the calibration process.

Optionally, when imagers are arranged in pairs, each imager is calibrated separately. For instance, when 8 imagers are used, the calibration process is executed on 4 even (left) imagers and then on 4 odd (right) imagers. The 0^thimager may be added artificially and temporarily to the odd imagers for calibration. This way, both even and odd imagers having a common field of view.

Reference is now made, once again, to FIG. 1. The figure describes a process wherein a pair of frames having an overlapping area depicting a common potion of the environment is processed for stitching the pair of frames. The process is optionally repeated, either iteratively or simultaneously, for pair of frames captured by neighboring imagers which are used to capture a set of frames. The set of frames optionally includes frames captured simultaneously or substantially simultaneously (e.g. with minor technical time drift) by a plurality of imagers mounted around a common center, for instance using the arrangements described above, for instance using the system depicted in FIG. 2 and optionally any of the arrangements depicted in FIG. 3A. Optionally, the methods is used for stitching multiple sets of frames into multiple panoramic frames, optionally stereoscopic, which are captured sequentially using an imaging device such as the imaging device depicted in FIG. 3B. This allows creating a VR file having a plurality of sequential panoramic frames for the creation and viewing of photographically-captured panoramas and the exploration of objects through frames taken at multiple viewing angles. The panoramic frame or image (referred to interchangeably) is optionally a VR panorama frame that documents the environment surrounding a center area to emulate an environment around a viewer (inside, looking out), yielding a sense of place and optionally changes in the place over time.

In each iteration, as shown at 102, a pair of frames having an overlapping area depicting a common potion of the environment is selected for processing from a set of frames which is captured simultaneously (a term used herein for describing also substantially simultaneously, for instance with a minor time deviation of less than 1 second). The set of frames comprises frames captured by imagers at multiple viewing angles, for example as described above, for instance using a system as depicted in FIG. 2 and/or an arrangement as depicted in FIG. 3A. Optionally the frames are projected on a sphere.

Now, as shown at 104, an overlapping area between the frames of the pair of frames is identified as described in co filed application titled “STITCHING FRAMES INTO A PANORAMIC FRAME” of the same inventors which is incorporated herein by reference (attorney reference number 64940). For further processing, a portion of one of the frames which depicts the frames is selected, optionally randomly.

Now, as shown at 105, irregular line path crossing the overlapping area along boundaries of the objects depicted in the overlapping area is identified in each pair of frames. Optionally, during this stage, feature points are detected in the respective overlapping area, for instance by executing a corner detection method, such as an accelerated segment test (FAST) or any other suitable feature detection process. For instance, see marked dots in FIG. 4 which are feature points identified using a FAST process. The feature points may be any point having one or more predefined feature point characteristic (value or a range of values), for instance a geometric shape and/or size, color, hue and/or any characteristic intensity.

Optionally, as shown at 106, objects depicted in the overlapping area, for instance areas or segments in the overlapping area, are detected, optionally based on the location of the feature points in the overlapping area. The objects may be detected using grow-region algorithms, such as a Watershed process, for growing regions around the feature points.

Now, as shown at 108, boundaries of the objects depicted in the overlapping area are identified, for instance using boundaries marking or identification processes.

This allows, as shown at 110, identifying an irregular line path crossing the overlapping area along the identified boundaries are selected along the boundaries, for instance the shortest irregular line path which crosses the overlapping area.

Optionally, the irregular line path is a path on vertexes marked on a graph built from the boundaries where G denotes the graph and G=(V, E, w) where V denotes intersections between the boundaries, for instance feature points indicative of corners, E denotes polylines between intersections, e denotes an edge of E, w(e) denotes a Sum of Avg_D (feature-point, polyline) for the nearest 10 features where Avg_D (point, polyline) denotes an average distance of point from points of polyline.

Optionally, the topmost point and bottommost vertexes are forced to be vertices through which the irregular line path. The graph is than analyzed for identifying the shortest irregular line path crossing the overlapping area along the boundaries of the segments or areas. For example, FIG. 4 is an exemplary overlapping area marked with feature points (dots), boundaries of objects (green colored lines), and a shortest irregular path which crosses the exemplary overlapping area along some of the boundaries, in accordance with some embodiments of the present invention. FIG. 5 depicts an exemplary overlapping area marked with boundaries of objects depicted in the exemplary overlapping area, in accordance with some embodiments of the present invention.

An example of stitched frames is provided in FIG. 6A, which is exemplary set of 4 frames which are captured by imagers at multiple viewing angles and FIGS. 6B-6C which are parts of a panoramic frame stitched in accordance with the process depicted in FIG. 1 and described above. The frames are acquired using system 200 as described with reference to FIG. 2.

Optionally, as shown at 115, in order to avoid rapid changes of content resulting in flickering when presenting one panoramic frame after the other, coordinates of the selected irregular path is matched with a previously calculated irregular path which is calculated for frames from the same imagers. When a shift higher than a predefined threshold is detected, another irregular path is calculated by shifting the coordinates of the previously calculated irregular path toward the coordinates of the current irregular path.

Optionally, as shown at 116, the respective frames are stitched along the irregular path (used as a seam). As shown at 118, the process is repeated for any pair of frames captured using neighboring imagers for creating a panoramic frame. As shown at 120, the created panoramic frame is added to a previously created panoramic frame (when applicable) for creating a panoramic VR file. For example see co filed application titled “STITCHING FRAMES INTO A PANORAMIC FRAME” of the same inventors which is incorporated herein by reference (attorney reference number 64940).

The descriptions of the various embodiments of the present invention have been presented for purposes of illustration, but are not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen to best explain the principles of the embodiments, the practical application or technical improvement over technologies found in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

It is expected that during the life of a patent maturing from this application many relevant imagers will be developed and the scope of the term frame is intended to include all such new technologies a priori.

As used herein the term “about” refers to ±10%.

The terms “comprises”, “comprising”, “includes”, “including”, “having” and their conjugates mean “including but not limited to”. This term encompasses the terms “consisting of” and “consisting essentially of”.

The phrase “consisting essentially of” means that the composition or method may include additional ingredients and/or steps, but only if the additional ingredients and/or steps do not materially alter the basic and novel characteristics of the claimed composition or method.

As used herein, the singular form “a”, “an” and “the” include plural references unless the context clearly dictates otherwise. For example, the term “a compound” or “at least one compound” may include a plurality of compounds, including mixtures thereof.

The word “exemplary” is used herein to mean “serving as an example, instance or illustration”. Any embodiment described as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments and/or to exclude the incorporation of features from other embodiments.

The word “optionally” is used herein to mean “is provided in some embodiments and not provided in other embodiments”. Any particular embodiment of the invention may include a plurality of “optional” features unless such features conflict.

Throughout this application, various embodiments of this invention may be presented in a range format. It should be understood that the description in range format is merely for convenience and brevity and should not be construed as an inflexible limitation on the scope of the invention. Accordingly, the description of a range should be considered to have specifically disclosed all the possible subranges as well as individual numerical values within that range. For example, description of a range such as from 1 to 6 should be considered to have specifically disclosed subranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6 etc., as well as individual numbers within that range, for example, 1, 2, 3, 4, 5, and 6. This applies regardless of the breadth of the range.

Whenever a numerical range is indicated herein, it is meant to include any cited numeral (fractional or integral) within the indicated range. The phrases “ranging/ranges between” a first indicate number and a second indicate number and “ranging/ranges from” a first indicate number “to” a second indicate number are used herein interchangeably and are meant to include the first and second indicated numbers and all the fractional and integral numerals therebetween.

It is appreciated that certain features of the invention, which are, for clarity, described in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the invention, which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable subcombination or as suitable in any other described embodiment of the invention. Certain features described in the context of various embodiments are not to be considered essential features of those embodiments, unless the embodiment is inoperative without those elements.

Although the invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art. Accordingly, it is intended to embrace all such alternatives, modifications and variations that fall within the spirit and broad scope of the appended claims.

All publications, patents and patent applications mentioned in this specification are herein incorporated in their entirety by reference into the specification, to the same extent as if each individual publication, patent or patent application was specifically and individually indicated to be incorporated herein by reference. In addition, citation or identification of any reference in this application shall not be construed as an admission that such reference is available as prior art to the present invention. To the extent that section headings are used, they should not be construed as necessarily limiting.

	Number	Date	Country
	62274321	Jan 2016	US
	62274317	Jan 2016	US

ADAPTIVE STITCHING OF FRAMES IN THE PROCESS OF CREATING A PANORAMIC FRAME

Information

Publication Number

Date Filed

Date Published

Inventors

Original Assignees

CPC

International Classifications

Abstract

Description

Claims

PCT Information

Provisional Applications (2)