Video perspective (VST) augmented reality (AR) device and method of operation thereof

Abstract

A method includes obtaining a stereoscopic image pair including a first image and a second image. The method further includes generating a first feature map of the first image and a second feature map of the second image, the first feature map and the second feature map including extraction locations associated with pixels in the images. The method further includes generating a disparity map between the first image and the second image based on the dense depth map. The method further includes generating a validated depth map based on a pixel-level comparison of predicted locations and extracted locations associated with at least some pixels in at least one of the images, the predicted locations determined based on the disparity map. Further, the method includes generating a first virtual view and a second virtual view based on the validated depth map for display on a display panel of the VST AR device.

Description

Technical Field

[0001] The present disclosure generally relates to augmented reality (AR) systems and processes. More specifically, the present disclosure relates to a system and method for parallax correction for video see-through (VST) AR. Background Art

[0002] Augmented reality (AR) systems can seamlessly blend virtual objects generated by computer graphics within real-world scenes. Optical see-through (OST) AR systems are AR systems in which the user directly views the real-world scene through a head-mounted device (HMD). Unfortunately, OST AR systems face many challenges that may limit their adoption. Some of these challenges include limited field of view, limited use spaces (such as indoor use only), inability to display completely opaque black objects, and the use of complex optical pipelines that may require projectors, waveguides, and other optical elements. Summary of the invention

[0003] The present disclosure provides a system and method for parallax correction for video see-through (VST) augmented reality (AR).

[0004] According to one aspect of the present disclosure, a method includes obtaining a stereoscopic image pair including a first image and a second image captured using a first perspective camera and a second perspective camera associated with a video perspective (VST) augmented reality (AR) device. The method also includes generating a first feature map of the first image and a second feature map of the second image, the first feature map including extraction positions associated with pixels in the first image, and the second feature map including extraction positions associated with pixels in the second image. The method also includes generating a disparity map between the first image and the second image based on a dense depth map. The method also includes generating a verified depth map based on a pixel-level comparison of predicted positions and extraction positions associated with at least some pixels in at least one of the first image and the second image, the predicted positions being determined based on the disparity map. In addition, the method includes generating a first virtual view and a second virtual view based on the verified depth map for display on a display panel of the VST AR device.

[0005] According to one aspect of the present disclosure, a VST AR device includes at least one display panel and a first perspective camera and a second perspective camera. The electronic device also includes at least one processing device, and the at least one processing device is configured to obtain a stereoscopic image pair including a first image and a second image captured using the first perspective camera and the second perspective camera. At least one processing device is also configured to generate a first feature map of the first image and a second feature map of the second image, the first feature map including extraction positions associated with pixels in the first image, and the second feature map including extraction positions associated with pixels in the second image. At least one processing device is also configured to generate a disparity map between the first image and the second image based on a dense depth map. At least one processing device is also configured to generate a verified depth map based on a pixel-level comparison of predicted positions and extraction positions associated with at least some pixels in at least one of the first image and the second image, and the predicted position is determined based on the disparity map. In addition, at least one processing device is configured to generate a first virtual view and a second virtual view based on the verified depth map to be displayed on the display panel of the VST AR device.

[0006] According to one aspect of the present disclosure, a computer-readable medium contains instructions that, when executed, cause at least one processor to perform the method.

[0007] Other technical features can be easily seen by those skilled in the art from the following drawings, descriptions and claims.

[0008] Before proceeding to the following detailed description, it may be advantageous to set forth definitions of specific words and phrases used throughout the patent document. The terms "send," "receive," and "communicate," and their derivatives, include both direct and indirect communications. The terms "include" and "comprises," and their derivatives, mean including but not limited to. The term "or" is inclusive, meaning and / or. The phrase "associated with," and its derivatives, means including, included within, interconnected with, containing, contained within, connected to or connected with, coupled to or coupled with, communicable with, cooperating with, interlaced, juxtaposed, proximate to, bound to or bound with, having, having the nature of, having a relationship with, and the like.

[0009] In addition, the various functions described below may be implemented or supported by one or more computer programs, each of which is composed of a computer-readable program code and embodied in a computer-readable medium. The terms "application" and "program" refer to one or more computer programs, software components, instruction sets, processes, functions, objects, classes, instances, related data, or a portion thereof, suitable for implementation in a suitable computer-readable program code. The phrase "computer-readable program code" includes any type of computer code, including source code, object code, and executable code. The phrase "computer-readable medium" includes any type of medium that can be accessed by a computer, such as a read-only memory (ROM), a random access memory (RAM), a hard drive, a compact disk (CD), a digital video disk (DVD), or any other type of memory. "Non-transitory" computer-readable media do not include wired, wireless, optical, or other communication links that transmit temporary electrical or other signals. Non-transitory computer-readable media include media that can permanently store data and media that can store data and rewrite it later, such as a rewritable optical disk or an erasable memory device.

[0010] As used herein, terms and phrases such as "having", "may have", "include" or "may include" a feature (such as a number, function, operation or component such as a part) indicate the presence of the feature and do not exclude the presence of other features. In addition, as used herein, the phrases "A or B", "at least one of A and / or B" or "one or more of A and / or B" may include all possible combinations of A and B. For example, "A or B", "at least one of A and B" and "at least one of A or B" may indicate all of the following: (1) including at least one A, (2) including at least one B, or (3) including at least one A and at least one B. In addition, as used herein, the terms "first" and "second" may modify various components regardless of importance and do not limit the components. These terms are only used to distinguish one component from another. For example, a first user device and a second user device may indicate user devices that are different from each other, regardless of the order or importance of the devices. Without departing from the scope of the present disclosure, a first component may be represented as a second component, and vice versa.

[0011] It should be understood that when an element (such as a first element) is referred to as being "coupled" / "coupled to" another element (such as a second element) or "connected" / "connected to" another element (such as a second element) (operably or communicatively), it may be directly coupled or connected / coupled to or connected to the other element, or coupled or connected to the other element / coupled to or connected to the other element via a third element. Conversely, it will be understood that when an element (such as a first element) is referred to as being "directly coupled" / "directly coupled to" another element (such as a second element) or "directly connected" / "directly connected to" another element (such as a second element) "with" another element (such as a second element), no other elements (such as a third element) are interposed between the element and the other element.

[0012] As used herein, the phrase "configured (or set) to" may be used interchangeably with the phrases "suitable for," "capable of," "designed to," "suitable for," "manufactured to," or "capable of," as appropriate. The phrase "configured (or set) to" does not inherently mean "specially designed in hardware to." On the contrary, the phrase "configured to" may indicate that a device may perform an operation together with another device or component. For example, the phrase "a processor configured (or set) to perform A, B, and C" may indicate a general-purpose processor (such as a CPU or an application processor) that can perform operations by running one or more software programs stored in a memory device, or a dedicated processor (such as an embedded processor) for performing operations.

[0013] The terms and phrases used herein are only provided for describing some embodiments of the present disclosure, rather than limiting the scope of other embodiments of the present disclosure. It should be understood that, unless the context clearly states otherwise, the singular form includes plural references. All terms and phrases used herein (including technical terms and phrases and scientific terms and phrases) have the same meaning as the meaning generally understood by those of ordinary skill in the art to which the embodiments of the present disclosure belong. It will also be understood that terms and phrases (such as those defined in a common dictionary) should be interpreted as having the same meaning as the meaning in the context of the relevant field, and unless clearly defined herein, will not be interpreted with an idealized or overly formal meaning. In some cases, the terms and phrases defined herein may be interpreted as not including embodiments of the present disclosure.

[0014] Examples of "electronic devices" according to embodiments of the present disclosure may include at least one of a smart phone, a tablet personal computer (PC), a mobile phone, a video phone, an e-book reader, a desktop PC, a laptop computer, a netbook computer, a workstation, a personal digital assistant (PDA), a portable multimedia player (PMP), an MP3 player, a mobile medical device, a camera, or a wearable device (such as smart glasses, a head mounted device (HMD), electronic clothing, an electronic bracelet, an electronic necklace, an electronic accessory, an electronic tattoo, a smart mirror, or a smart watch). Other examples of electronic devices include smart home appliances. Examples of smart home appliances may include a television, a digital video disc (DVD) player, an audio player, a refrigerator, an air conditioner, a vacuum cleaner, an oven, a microwave, a washing machine, a dryer, an air purifier, a set-top box, a home automation control panel, a security control panel, a TV box (such as SAMSUNG HOMESYNC, APPLETV, or GOOGLE TV), a smart speaker or a speaker with an integrated digital assistant (such as SAMSUNG GALAXY HOME, APPLE HOMEPOD, or AMAZON ECHO), a game console (such as XBOX, PLAYSTATION, or NINTENDO), an electronic dictionary, an electronic key, a camera, or at least one of an electronic photo frame. Other examples of electronic devices include various medical devices (such as various portable medical measuring devices (such as blood sugar measuring devices, heart rate measuring devices, or body temperature measuring devices), magnetic resonance angiography (MRA) devices, magnetic resonance imaging (MRI) devices, computed tomography (CT) devices, imaging devices, or ultrasound devices), navigation devices, global positioning system (GPS) receivers, event data recorders (EDR), flight data recorders (FDR), car infotainment devices, marine electronic devices (such as marine navigation devices or gyrocompasses), avionics devices, security devices, vehicle-mounted host units, industrial or domestic robots, automatic teller machines (ATMs), point-of-sale (POS) devices, or Internet of Things (IoT) devices (such as light bulbs, various sensors, electric or gas meters, sprinklers, fire alarms, thermostats, street lights, ovens, fitness devices, hot water tanks, heaters, or boilers). Other examples of electronic devices include at least a portion of a piece of furniture or a building / structure, an electronic board, an electronic signature receiving device, a projector, or various measuring devices (such as devices for measuring water, electricity, gas, or electromagnetic waves). Note that according to various embodiments of the present disclosure, the electronic device may be one or a combination of the devices listed above. According to some embodiments of the present disclosure, the electronic device may be a flexible electronic device. The electronic device disclosed herein is not limited to the devices listed above, and may include new electronic devices according to the development of technology.

[0015] In the following description, an electronic device is described with reference to the accompanying drawings according to various embodiments of the present disclosure. As used herein, the term "user" may refer to a person using an electronic device or another device (such as an artificial intelligence electronic device).

[0016] Definitions for additional specific words and phrases may be provided throughout this patent document. Those of ordinary skill in the art should understand that in many, if not most instances, such definitions apply to prior, as well as future uses of such defined words and phrases.

[0017] None of the description in this application should be read as implying that any particular element, step, or function is an essential element that must be included in the claims scope. The scope of the patented subject matter is limited solely by the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] For a more complete understanding of the present disclosure and its advantages, reference is now made to the following description taken in conjunction with the accompanying drawings, wherein like reference numerals represent like parts:

[0019] Figure 1 An example network configuration including electronic devices according to the present disclosure is shown;

[0020] Figure 2 An example process for parallax correction for video see-through augmented reality according to the present disclosure is shown;

[0021] Figure 3A shows an example arrangement between a perspective camera viewpoint and an eye (virtual camera) viewpoint according to the present disclosure;

[0022] Figure 3B shows an example arrangement between a perspective camera viewpoint and an eye (virtual camera) viewpoint according to the present disclosure;

[0023] Figure 4 shows further details of the in-depth clarification and verification process according to the present disclosure;

[0024] Figure 5 Details of depth correction and noise reduction operations according to the present disclosure are shown;

[0025] Figure 6 An example of consistency between images according to the present disclosure is shown;

[0026] Figure 7 shows an example graph showing correct depth information and incorrect depth information according to the present disclosure;

[0027] Figure 8 shows an example diagram showing further details of a left virtual view and a right virtual view according to the present disclosure;

[0028] Fig. 9 Details of a process for generating parallax-corrected virtual objects through a graphics pipeline for display on a VST AR head mounted device according to the present disclosure are shown; and

[0029] Fig.10 An example method for parallax correction for video see-through augmented reality according to the present disclosure is shown. DETAILED DESCRIPTION

[0030] The following discussion is described with reference to the accompanying drawings. Figures 1 to 10 However, it should be understood that the present disclosure is not limited to these embodiments, and all changes and / or equivalent forms or replacement forms thereof also belong to the scope of the present disclosure.

[0031] In various embodiments, the perspective camera is typically a high-resolution camera (such as a 2K or 4K camera or higher resolution camera). In order to provide a quality user experience with an AR headset, the latency of the video frame transformation may need to be as low as possible to reduce or prevent the user from noticing the latency when moving their head. However, existing techniques are generally unable to process high-resolution images from a perspective camera to generate a virtual view frame with sufficiently low latency.

[0032] Unlike optical see-through AR, where the user can directly see the surrounding scene, video see-through (VST) AR uses a see-through camera mounted on an AR head-mounted device to reconstruct the surrounding scene. Because the position of the see-through camera is different from the position of the user's eyes, a virtual image at the viewpoint of the user's eyes is generated from the image frames captured at the viewpoint of the see-through camera. In this way, the user's eyes can see the external scene through the camera, just as if the see-through camera was mounted at the viewpoint of the eye.

[0033] VST AR has some advantages over other types of AR, including wider field of view, usability in outdoor environments, dark occlusion, and altered perception. However, VST AR systems also face certain challenges. For example, in VST AR, since the perspective camera cannot be installed in the same position of the eye, the perspective camera viewpoint is different from the eye viewpoint. Due to the difference in viewpoint, the disparity map from the perspective camera viewpoint is different from the disparity map from the eye viewpoint. In order to obtain the correct view at the virtual camera viewpoint, the view from the perspective camera needs to be transformed to a virtual camera with parallax correction.

[0034] The present disclosure provides various techniques for parallax correction for video perspective augmented reality. As described in more detail below, the disclosed system and method provide an efficient algorithm and pipeline for generating a correct view at a virtual camera viewpoint based on a perspective camera view through parallax correction. The perspective camera and the virtual camera are located at different positions of the HMD. Note that although some of the embodiments discussed below are described in the context of use of a consumer electronic device (such as an AR head-mounted device), this is merely an example, and it will be understood that the principles of the present disclosure can be implemented in many different other suitable environments, and any suitable device can be used.

[0035] Figure 1 An example network configuration 100 including electronic devices according to the present disclosure is shown. Figure 1 The embodiment of the network configuration 100 shown in FIG. 1 is for illustration only. Other embodiments of the network configuration 100 may be used without departing from the scope of the present disclosure.

[0036] According to an embodiment of the present disclosure, an electronic device 101 is included in a network configuration 100. The electronic device 101 may include at least one of a bus 110, a processor 120, a memory 130, an input / output (I / O) interface 150, a display 160, a communication interface 170, or a sensor 180. In some embodiments, the electronic device 101 may not include at least one of these components, or may add at least one other component. The bus 110 includes circuits for connecting components 120-180 to each other and for transmitting communications (such as control messages and / or data) between components.

[0037] The processor 120 includes one or more processing devices, such as one or more microprocessors, microcontrollers, digital signal processors (DSPs), application specific integrated circuits (ASICs), or field programmable gate arrays (FPGAs). In some embodiments, the processor 120 includes one or more of a central processing unit (CPU), an application processor (AP), a communication processor (CP), or a graphics processor unit (GPU). The processor 120 is capable of performing control on at least one of the other components of the electronic device 101 and / or performing operations or data processing related to communication or other functions. As described in more detail below, the processor 120 may perform one or more operations for parallax correction for video perspective augmented reality.

[0038] The memory 130 may include a volatile memory and / or a non-volatile memory. For example, the memory 130 may store commands or data related to at least one other component of the electronic device 101. According to an embodiment of the present disclosure, the memory 130 may store software and / or a program 140. The program 140 includes, for example, a core 141, a middleware 143, an application programming interface (API) 145, and / or an application program (or "application") 147. At least a portion of the core 141, the middleware 143, or the API 145 may be represented as an operating system (OS).

[0039] The core 141 may control or manage system resources (such as the bus 110, the processor 120, or the memory 130) for executing operations or functions implemented in other programs (such as the middleware 143, the API 145, or the application 147). The core 141 provides an interface that allows the middleware 143, the API 145, or the application 147 to access various components of the electronic device 101 to control or manage system resources. As described below, the application 147 may support one or more functions for parallax correction for video perspective augmented reality. These functions may be performed by a single application or by multiple applications that respectively perform one or more of these functions.

[0040] For example, the middleware 143 may be used as a repeater for allowing the API 145 or the application 147 to transmit data with the core 141. A plurality of applications 147 may be provided. The middleware 143 is capable of controlling work requests received from the applications 147, such as by assigning a priority for using system resources (such as the bus 110, the processor 120, or the memory 130) of the electronic device 101 to at least one of the plurality of applications 147. The API 145 is an interface that allows the application 147 to control functions provided from the core 141 or the middleware 143. For example, the API 145 includes at least one interface or function (such as a command) for filing control, window control, image processing, or text control.

[0041] The input / output (I / O) interface 150 serves as an interface that can, for example, transmit commands or data input from a user or other external devices to other components of the electronic device 101. The I / O interface 150 can also output commands or data received from other components of the electronic device 101 to the user or other external devices.

[0042] Display 160 includes, for example, a liquid crystal display (LCD), a light emitting diode (LED) display, an organic light emitting diode (OLED) display, a quantum dot light emitting diode (QLED) display, a micro-electromechanical system (MEMS) display, or an electronic paper display. Display 160 may also be a depth perception display, such as a multi-focus display. Display 160 is capable of displaying, for example, various contents (such as text, images, videos, icons, or symbols) to a user. Display 160 may include a touch screen, and may receive, for example, a touch, gesture, proximity, or hover input using an electronic pen or a body part of a user. Display 160 may also refer to a display panel.

[0043] For example, the communication interface 170 can establish communication between the electronic device 101 and an external electronic device (such as the first external electronic device 102, the second external electronic device 104, or the server 106). For example, the communication interface 170 can be connected to the network 162 or 164 through wireless communication or wired communication to communicate with the external electronic device. The communication interface 170 can be a wired transceiver or a wireless transceiver or any other component for sending and receiving signals.

[0044] Wireless communication can use, for example, long term evolution (LTE), long term evolution-advanced (LTE-A), 5th generation wireless system (5G), millimeter wave or 60GHz wireless communication, wireless USB, code division multiple access (CDMA), wideband code division multiple access (WCDMA), universal mobile telecommunications system (UMTS), wireless broadband (WiBro) or global system for mobile communications (GSM) as at least one of the cellular communication protocol. Wired connection may include, for example, at least one of universal serial bus (USB), high-definition multimedia interface (HDMI), recommended standard 232 (RS-232) or plain old telephone service (POTS). Network 162 or 164 includes at least one communication network, such as a computer network (such as a local area network (LAN) or a wide area network (WAN)), the Internet or a telephone network.

[0045] The electronic device 101 also includes one or more sensors 180, which can measure physical quantities or detect the activation state of the electronic device 101, and convert the measured or detected information into electrical signals. For example, one or more sensors 180 include one or more cameras or other image sensors (such as perspective cameras) for capturing images of scenes. The sensor 180 may also include one or more buttons for touch input, gesture sensors, gyroscopes or gyroscope sensors, atmospheric pressure sensors, magnetic sensors or magnetometers, acceleration sensors or accelerometers, grip sensors, proximity sensors, color sensors (such as red, green, and blue (RGB) sensors), biophysical sensors, temperature sensors, humidity sensors, illumination sensors, ultraviolet (UV) sensors, electromyography (EMG) sensors, electroencephalography (EEG) sensors, electrocardiography (ECG) sensors, infrared (IR) sensors, ultrasonic sensors, iris sensors, or fingerprint sensors. The sensor 180 may also include an inertial measurement unit, which may include one or more accelerometers, gyroscopes, and other components. In addition, the sensor 180 may include a control circuit for controlling at least one of the sensors included herein. Any of these sensors 180 may be located within the electronic device 101 .

[0046] In some embodiments, the electronic device 101 may be a wearable device or a wearable device (such as an HMD) on which the electronic device can be installed. For example, the electronic device 101 may represent an AR wearable device, such as a head-mounted device with a display panel or smart glasses. In other embodiments, the first external electronic device 102 or the second external electronic device 104 may be a wearable device or a wearable device (such as an HMD) on which the electronic device can be installed. In those other embodiments, when the electronic device 101 is installed in the first external electronic device 102 (such as an HMD), the electronic device 101 may communicate with the first external electronic device 102 through the communication interface 170. The electronic device 101 may be directly connected to the first external electronic device 102 to communicate with the first external electronic device 102 without involving a separate network.

[0047] The first external electronic device 102 and the second external electronic device 104 and the server 106 may be devices of the same or different types as the electronic device 101, respectively. According to a specific embodiment of the present disclosure, the server 106 includes a group of one or more servers. In addition, according to a specific embodiment of the present disclosure, all operations or some operations performed on the electronic device 101 may be performed on another electronic device or multiple other electronic devices (such as the first external electronic devices 102 and 104 or the server 106). In addition, according to a specific embodiment of the present disclosure, when the electronic device 101 should automatically or by request perform a certain function or service, the electronic device 101 may request another device (such as the first external electronic device 102 and the second external electronic device 104 or the server 106) to perform at least some functions associated with it, instead of performing the function or service separately or additionally. Another electronic device (such as the first external electronic device 102 and the second external electronic device 104 or the server 106) is capable of performing the requested function or additional function and transmitting the result of the execution to the electronic device 101. The electronic device 101 can provide the requested function or service by processing the received result as is or additionally. To this end, for example, cloud computing, distributed computing, or client-server computing techniques may be used. Figure 1 The electronic device 101 is shown to include a communication interface 170 for communicating with a second external electronic device 104 or a server 106 via a network 162 or 164 , but according to some embodiments of the present disclosure, the electronic device 101 may operate independently without a separate communication function.

[0048] The server 106 may include components 110-180 (or a suitable subset thereof) that are the same or similar to the electronic device 101. The server 106 may support driving the electronic device 101 by performing at least one of the operations (or functions) implemented on the electronic device 101. For example, the server 106 may include a processing module or processor that may support the processor 120 implemented in the electronic device 101. As described in more detail below, the server 106 may perform one or more operations for supporting techniques for parallax correction for video see-through augmented reality.

[0049] although Figure 1 An example of a network configuration 100 including an electronic device 101 is shown, but the Figure 1 Various changes may be made. For example, network configuration 100 may include any number of each component in any suitable arrangement. In general, computing systems and communication systems have a variety of configurations, and Figure 1 The scope of the present disclosure is not limited to any particular configuration. Figure 1One operating environment is shown in which the various features disclosed in this patent document may be used, but these features may be used in any other suitable system.

[0050] Figure 2 An example process 200 for parallax correction for video see-through augmented reality according to the present disclosure is shown. For ease of explanation, the process 200 is described as using the above Figure 1 The process 200 may be performed by one or more components of the network configuration 100, such as the electronic device 101. However, this is merely an example, and the process 200 may be performed using any other suitable device and in any other suitable system.

[0051] like Figure 1 and Figure 2 As shown, the electronic device 101 includes a plurality of sensors 180 , and the sensors 180 include one or more video perspective cameras 201 , one or more gesture tracking cameras 202 , one or more depth sensors 203 , and one or more position sensors 204 .

[0052] The video perspective camera 201 may include a first high-resolution perspective camera and a second high-resolution perspective camera 201. In some embodiments, the video perspective camera 201 may be arranged as a left high-resolution perspective camera and a right high-resolution perspective camera. Depending on the capabilities of the video perspective camera 201, the video perspective camera 201 may capture an image 206 having any suitable resolution and size. In some embodiments, for example, the image 206 includes high-resolution RGB image data, which typically includes image data in three color channels (i.e., a red channel, a green channel, and a blue channel). However, the image 206 may include image data having any other suitable form or arrangement. As described in more detail below, the image 206 may include a left image and a right image, and may be used to generate a final virtual view and render it on a panel of the HMD.

[0053] The gesture tracking camera 202 may include any suitable camera capable of capturing information for HMD gesture tracking. For example, the gesture tracking camera 202 may capture HMD image gesture tracking information, which may be used with the left image and the right image 206 to generate a stereoscopic image pair. In some embodiments, the gesture tracking camera 202 may include a left gesture tracking camera and a right gesture tracking camera. However, other numbers and arrangements of the gesture tracking cameras 202 are possible.

[0054] The depth sensor 203 may include any suitable type of depth sensor, such as a ToF sensor. The depth sensor 203 may capture depth data 207 for depth reprojection and parallax correction. The position sensor 204 may include any suitable type of position sensor, such as an IMU sensor. The position sensor 204 may capture head position data 208 of a user wearing the HMD. The electronic device 101 may include any suitable number and arrangement of the depth sensors 203 and the position sensors 204.

[0055] Figure 3A An example arrangement 300 between a perspective camera viewpoint and an eye (virtual camera) viewpoint according to the present disclosure is shown. Figure 3B An example arrangement 305 between a perspective camera viewpoint and an eye (virtual camera) viewpoint according to the present disclosure is shown.

[0056] like Figure 3A and Figure 3B As shown, arrangement 300 and arrangement 305 illustrate examples of parallax due to the relationship between a 3D scene, perspective camera imaging, and virtual camera imaging. Figure 3A and Figure 3B As shown, two perspective cameras 302 and 303 are arranged as a left perspective camera and a right perspective camera. Perspective camera 302 and perspective camera 303 may represent (or be represented by) video perspective camera 201. Figure 3A Also shown are the positions of left and right virtual cameras 312 and 313, which may represent the viewpoints seen from the left and right eyes of a user of the HMD.

[0057] exist Figure 3A , the positional differences between the right perspective camera 303 and the right virtual camera 313 include dx in the X direction, dy in the Y direction, and dz in the Z direction. A similar positional difference exists between the left perspective camera 302 and the left virtual camera 312. When looking at a point 322 (identified as a) on a 3D object 320 of the scene, the left perspective camera 302 sees point a 1 , while the left virtual camera 312 sees point a 2 The difference in views is the result of parallax.

[0058] exist Figure 3B , the position of the left perspective camera 302 has changed relative to the left virtual camera 312, so that the position difference between the left perspective camera 302 and the left virtual camera 312 only includes dz in the Z direction (i.e., for the left perspective camera 302 and the left virtual camera 312, Figure 3B Although there is still some parallax, Figure 3B Point a seen in 1 and a 2 Than in Figure 3A Closer together.

[0059] Go to again Figure 2 At operation 205, the electronic device 101 obtains one or more camera parameters and one or more distortion models of the video perspective camera 201. The camera parameters may include any suitable number of any suitable parameters, such as intrinsic parameters (e.g., focal length, lens distortion, etc.) and extrinsic parameters (e.g., transformation parameters, etc.). The distortion model may include any suitable distortion model representing the distortion of the video perspective camera 201. At operation 209, the electronic device 101 dedistorts and corrects the image 206 using the perspective camera parameters and the distortion model to obtain a left corrected image 214 and a right corrected image 215. At operation 210, the electronic device 101 inputs the corrected image 214 and the corrected image 215 into a deep neural network (DNN) to reconstruct a depth map of the image 206. The DNN represents any suitable trained or untrained neural network configured to generate a depth map from an image.

[0060] At operation 211, the electronic device 101 performs gesture tracking using the HMD image gesture tracking information captured by the gesture tracking camera 202 and the head position data 208 captured by the position sensor 204. By performing gesture tracking, the electronic device 101 obtains a six-degree-of-freedom (6DoF) camera pose. At operation 212, the electronic device 101 performs a SLAM process using the 6DoF camera pose to generate sparse depth points.

[0061] At operation 213, the electronic device 101 performs depth data fusion to integrate depth data from the reconstructed depth map output by the DNN, the depth data 207 captured by the depth sensor 203, and the sparse depth points generated by the SLAM process in operation 212. The results of operation 213 include the creation of a dense depth map and a corresponding confidence map, which can be provided as input to the depth clarification and verification process 230.

[0062] The electronic device 101 uses the dense depth map generated in operation 213 to perform a depth clarification and verification process 230 to obtain a high-quality verified depth map 218. The verified depth map 218 will be used for disparity correction later in the process 200 to generate correct left and right virtual views at the eye viewpoint. As described in more detail below, the depth clarification and verification process 230 includes an effective color image guided algorithm to correct inaccurate depth and remove noise.

[0063] As an overview, the depth clarification and verification process 230 includes the following operations. At operation 231, the electronic device 101 extracts image features from the stereo image pair 206, and at operation 232, the electronic device 101 calculates a disparity map based on the dense depth map. At operation 233, using the image features and the disparity map of the left image 206, the electronic device 101 calculates corresponding image features of the right image 206. At operation 234, the electronic device 101 compares the extracted image features of the right image 206 (obtained in operation 231) with the calculated image features of the right image 206 (determined in operation 233) to determine the correspondence between the points in each image feature. At operation 235, if the calculated features and the extracted features sufficiently correspond, the electronic device 101 stores the depth, or if the calculated features and the extracted features do not sufficiently correspond, the electronic device 101 corrects the depth to obtain a more accurate depth. This will now be described in conjunction with Figure 4 Describe the process in more detail.

[0064] Figure 4 2 shows further details of the in-depth clarification and verification process 230 according to the present disclosure. Figure 4 As shown, Figure 4 Some of the operations shown and described in Figure 2 , and will be explained in this case. Figure 4 As shown, the inputs to the depth clarification and verification process 230 include the stereo image pair 206 from the perspective camera 201, the depth map 402 reconstructed using the DNN in operation 210, the depth data 207 captured by the depth sensor 203, and the sparse depth points and 6DoF poses 404 obtained from SLAM pose tracking in operations 211 and 212.

[0065] As previously discussed, at operation 209, the electronic device 101 dedistorts and corrects the stereo image pair 206 (using the perspective camera parameters and the distortion model from operation 205) to obtain a left corrected image 214 and a right corrected image 215. In addition, at operation 213, the electronic device 101 performs depth data fusion to integrate depth data from the reconstructed depth map output by the DNN, the depth data 207 captured by the depth sensor 203, and the sparse depth points 404 to obtain a dense depth map and a corresponding confidence map.

[0066] At operation 406, the electronic device 101 performs feature detection and extraction on the left corrected image 214 and the right corrected image 215 output from operation 209. At operation 407, the electronic device 101 generates a feature map of the left corrected image 214 based on the extracted features. The feature map of the left corrected image 214 includes extraction locations associated with pixels in the left corrected image 214. At operation 408, the electronic device 101 generates a feature map of the right corrected image 215 based on the extracted features. The feature map of the right corrected image 215 includes extraction locations associated with pixels in the right corrected image 215. Any suitable technique or algorithm may be used to generate the feature maps of the left corrected image 214 and the right corrected image 215. Operations 406-408 may represent or include Figure 2 Operation 231.

[0067] At operation 410, the electronic device 101 creates a disparity map between the left corrected image 214 and the right corrected image 215 using the dense depth map generated in operation 213. At operation 411, the electronic device 101 calculates a feature map of the right corrected image 215 using the disparity map generated in operation 410 and the feature map of the left corrected image 214. The electronic device 101 may use any suitable technique to create the disparity map and the feature map here.

[0068] At operation 412, the electronic device 101 compares the extracted feature map of the right corrected image 215 (generated in operation 408) with the calculated feature map of the right corrected image 215 (calculated in operation 411). The electronic device 101 performs the comparison using a pixel-by-pixel comparison with a predetermined threshold. At operation 414, the electronic device 101 determines whether the difference between the current pixel of the extracted feature map and the corresponding pixel of the calculated feature map is less than a predetermined threshold. If the difference between the pixel of the extracted feature map and the corresponding pixel of the calculated feature map is less than the predetermined threshold, then at operation 415, the electronic device 101 stores the depth of the pixel to the final depth map to be used later. Otherwise, if the difference between the pixel of the extracted feature map and the corresponding pixel of the calculated feature map is equal to or greater than the predetermined threshold, the electronic device 101 performs a depth correction and noise reduction operation 420 on the depth of the current pixel. In the depth correction and noise reduction operation 420, the electronic device 101 performs a color image-guided depth correction. Here, multiple types of information are used for depth correction, including depth information, color texture information, and spatial information.

[0069] Figure 5 Details of a depth correction and noise reduction operation 420 in accordance with the present disclosure are shown.

[0070] like Figure 5 As shown, Figure 5Detail of a depth correction and noise reduction operation 420 using depth information, color texture information, and spatial information according to the present disclosure is shown. Figure 5 As shown, at operation 501, the electronic device 101 calculates the current pixel p and the adjacent pixel p of the extracted feature map. nn ∈N(p) is the normalized depth difference D between norm , such as calculated by the following equation:

[0071] D norm =||d d (p)-d d (p nn )||, (1)

[0072] Among them, d d (p) is the depth at pixel p, and d d (p nn ) is the neighboring pixel p nn The depth at which p nn ∈N(p).

[0073] At operation 502, the electronic device 101 calculates the depth difference D according to the normalized depth difference D norm Calculate the weight W d , such as calculated by the following equation:

[0074]

[0075] Among them, D norm is the normalized depth difference, and (σ d , μ d ) are the standard deviation and mean in a Gaussian distribution.

[0076] In operation 503, the electronic device 101 calculates the current pixel p and the adjacent pixel p nn ∈N(p) is the normalized color texture difference T between norm , such as calculated by the following equation:

[0077] T norm =||d c (p)-d c (p nn )||, (3)

[0078] Among them, d c (p) is the color at pixel p, and d c (p nn ) is the neighboring pixel p nn The color of the place.

[0079] At operation 504, the electronic device 101 calculates the color texture difference T according to the normalized color texture difference T normCalculate the weight W c ,

[0080]

[0081] Among them, T norm is the normalized color-texture difference, and (σ c , μ c ) are the standard deviation and mean in a Gaussian distribution.

[0082] In operation 505, the electronic device 101 calculates the current pixel p and the adjacent pixel p nn ∈N(p) is the normalized spatial difference S between norm , such as calculated by the following equation:

[0083] S norm =||d s (p)-d s (p nn )||, (5)

[0084] Among them, d s (p) is the position at pixel p, and d s (p nn ) is the neighboring pixel p nn The location of the place.

[0085] At operation 506, the electronic device 101 calculates the normalized spatial difference S norm Calculate the weight W s , such as calculated by the following equation:

[0086]

[0087] Among them, S norm is the normalized spatial difference, and (σ s , μ s ) are the standard deviation and mean in a Gaussian distribution.

[0088] At operation 507, the electronic device 101 calculates the depth d(p) at the considered pixel p using a weighted average of the depth at the neighboring pixels, the color at the neighboring pixels, and the position at the neighboring pixels, such as by the following equation:

[0089]

[0090] Go to again Figure 4, after the electronic device 101 performs the depth correction and noise reduction operation 420 to obtain the depth d(p), the electronic device 101 repeats the operations 411, 412, and 414 until the electronic device 101 obtains a better depth. The electronic device 101 performs these operations on all pixels in the extracted feature map and the corresponding calculated feature map to obtain a verified depth map 218. The verified depth map 218 helps to ensure consistency between the left image and the right image 206 captured by the perspective camera 201.

[0091] Figure 6 An example 600 of consistency between images according to the present disclosure is shown. Figure 6 As shown, a point 601 on an object (identified as a) appears in views 604 and 605 of two perspective cameras 602 and 603 (which may represent perspective camera 201). Point a in views 604 and 605 1 and point a 2 are consistent because they all appear on the straight line between perspective camera 602 and perspective camera 603 and point 601. However, point a 1 and point a′ 2 is inconsistent, because point a′ 2 With point a 2 Not on the same line.

[0092] Figure 7 An example graph showing correct depth information and incorrect depth information according to the present disclosure is shown. Figure 7 As shown, graph 700 shows depth information of a plurality of pixels. 2 At the right depth And pixel a′ 2 At incorrect depth or in noise Depth correction and noise reduction operation 420 generates the correct depth using image-guided depth reconstruction and replaces the incorrect depth with the reconstructed correct depth.

[0093] Go to again Figure 2 After the electronic device 101 generates the verified depth map 218 in the depth clarification and verification process 230, the electronic device 101 uses the verified depth map 218, the left and right corrected images 214 and 215, and the 6DoF poses to generate the left virtual view 216 and the right virtual view 217. The left virtual view 216 and the right virtual view 217 feature parallax correction from the perspective camera viewpoint to the virtual camera viewpoint.

[0094] Figure 8 An example diagram 800 showing further details of the left virtual view 216 and the right virtual view 217 according to the present disclosure is shown. Figure 2 and Figure 8 As shown, when the left virtual camera 312 and the right virtual camera 313 view the same point P(x, d) on the object 320, the left virtual camera 312 sees the point p in the image plane. l (x l , f), the right virtual camera 313 sees point p in the image plane r (x r , f).

[0095] The relationship between the left virtual view 216 and the right virtual view 217 is given as follows:

[0096]

[0097] Among them, (x l , f) is the pixel position on the left virtual view 216, (x r , f) is the pixel position on the right virtual view 217, B is the distance between the left virtual camera 312 and the right virtual camera 313, f is the focal length of the virtual camera 312 and the virtual camera 313, and d is the depth of the point P(x, d) obtained from the verified depth map 218.

[0098] Therefore, the electronic device 101 can use the following equation according to the left virtual view 216 (x l ) and the verified depth map 218 to calculate the right virtual view 217 (x r ):

[0099]

[0100] In addition, the electronic device 101 performs a warping operation 219 to warp the verified depth map 218 into a left warped depth map 220 and a right warped depth map 221 corresponding to the left virtual camera 312 and the right virtual camera 313. At operation 224, the electronic device 101 creates a virtual object using the graphics pipeline and the left warped depth map 220 and the right warped depth map 221. In this way, the created virtual object has been parallax-corrected for the left virtual view 216 and the right virtual view 217.

[0101] At operation 225, the electronic device 101 mixes the real-world information 222 and 223 captured from the perspective camera 201 with the virtual information (objects) generated by the graphics pipeline in operation 224 and the warped depth maps 220 and 221 to create a final (parallax-corrected) left virtual view 226 and a final (parallax-corrected) right virtual view 227, so that the electronic device 101 can display the integrated information with a correct view at the eye viewpoint 228 and the eye viewpoint 229 in the VST AR head-mounted device.

[0102] Fig. 9 9 shows further details of a process 900 for generating parallax-corrected virtual objects through a graphics pipeline for display on a VST AR head mounted device in accordance with the present disclosure. Fig. 9 As shown, many of the operations shown and described in process 900 represent Figure 2 , and will be explained in this case.

[0103] The electronic device 101 obtains real-world depth information including the left warped depth map 220 and a dense disparity map 902 generated from the left warped depth map 220. Then, the electronic device 101 generates a virtual object using a graphics pipeline 920, which may be, for example, OpenGL, Unity, etc. This may be performed with Figure 2 The electronic device 101 generates a left virtual object 921 in the left virtual view using the left warped depth map 220, so that the left virtual object 921 has the same disparity as the left virtual view 931 generated from the perspective camera 201. The electronic device 101 generates a right virtual object 922 in the right virtual view using the generated left virtual object 921 and the dense disparity map 902. In this way, the generated right virtual object 922 has the same disparity as the right virtual view 932 generated from the perspective camera 201. The left virtual view 931 and the right virtual view 932 are from the perspective camera 201 or other sensors, and the disparity is corrected using the eye viewpoint 228 and the eye viewpoint 229.

[0104] At operation 923, the electronic device 101 blends the parallax-corrected left virtual object 921 and the right virtual object 922 with the left virtual view 931 and the right virtual view 932 by considering the occlusion relationship with the left warped depth map 220. Figure 2 The electronic device 101 then obtains a final left virtual view 226 in which both the virtual object and the real world information are mixed and corrected for parallax, and obtains a final right virtual view 227 in which both the virtual object and the real world information are mixed and corrected for parallax.

[0105] Finally, the electronic device 101 renders and displays the final left virtual view 226 on the left panel 926 of the VST AR head-mounted device, and renders and displays the final right virtual view 227 on the right panel 927 of the VST AR head-mounted device. The user can see information at the eye viewpoint 228 and the eye viewpoint 229.

[0106] although Figures 2 to 9 An example of a process 200 and related details for parallax correction for video see-through augmented reality is shown, but may be used for Figures 2 to 9For example, although described as involving a specific order of operations, Figures 2 to 9 The various operations of the described techniques may overlap, occur in parallel, occur in a different order, or occur any number of times (including zero). Figures 2 to 9 The specific operations shown in are examples only and may be performed using other techniques. Figures 2 to 9 Each operation in the operations shown in .

[0107] As an example of another technique, the electronic device 101 uses the relationship of a stereo camera pair to clarify and verify the depth integrated from various sources. The electronic device 101 can extend the technique using other techniques for depth clarification and verification. For example, the electronic device 101 can create an energy function with a combination of data terms and smooth terms. The data term describes the depth difference between the point under consideration and a point with high confidence. The smooth term describes the depth difference between the point under consideration and an adjacent point. By minimizing the energy function, the electronic device 101 can keep the depth at points with high confidence unchanged and make other points closer to adjacent values ​​without mutations. Using this technique, the electronic device 101 can improve the depth map generated by the previous steps. Using the improved depth map, the electronic device 101 can create a left virtual view and a right virtual view with corrected parallax.

[0108] As another example, the above techniques are characterized by a system with two perspective cameras. These techniques can be extended to systems with more than two perspective cameras. For example, the electronic device 101 can reconstruct depth maps from these multiple cameras and create left and right virtual views at the virtual camera viewpoints by reprojecting the multiple perspective cameras to the virtual camera. The electronic device 101 uses the depth maps obtained from the multiple perspective cameras to create a virtual object with parallax corrected as described above. The electronic device 101 then integrates the real world information captured from the perspective cameras to create the final virtual view.

[0109] Note that the above method can be implemented in any suitable manner in the electronic device 101, the first external electronic device 102 and the second external electronic device 104, the server 106 or other devices. Figures 2 to 9 shown in or about Figures 2 to 9 For example, in some embodiments, one or more software applications or other software instructions executed by the processor 120 of the electronic device 101, the first external electronic device 102 and the second external electronic device 104, the server 106 or other devices may be used to implement or support the operation and functions described in the embodiment of the present invention. Figures 2 to 9 shown in or about Figures 2 to 9 In other embodiments, dedicated hardware components may be used to implement or support the operations and functions described in Figures 2 to 9 shown in or about Figures 2 to 9In general, any suitable hardware or any suitable combination of hardware and software / firmware instructions may be used to implement at least some of the operations and functions described in the Figures 2 to 9 shown in or about Figures 2 to 9 Describes the operation and functionality.

[0110] Fig.10 An example method 1000 for parallax correction for video see-through augmented reality according to the present disclosure is shown. For ease of explanation, Fig.10 The method 1000 shown in FIG. 1 is described as involving using Figure 2 The process 200 and Figure 1 However, Fig.10 The method 1000 shown in FIG. 1 may be used with any other suitable processes and apparatus.

[0111] like Figure 2 and Fig.10 As shown, at step 1001, a stereoscopic image pair including a first image and a second image captured using a first perspective camera and a second perspective camera associated with a VS AR device is obtained. This may include, for example, the electronic device 101 using a left perspective camera and a right perspective camera 201 to obtain a left image and a right image 206.

[0112] At step 1003, a first feature map of the first image and a second feature map of the second image are generated. The first feature map includes extraction locations associated with pixels in the first image, and the second feature map includes extraction locations associated with pixels in the second image. This may include, for example, the electronic device 101 performing operations 407 and 408 to generate a left feature map and a right feature map from the left corrected image 214 and the right corrected image 215.

[0113] At step 1005, a dense depth map is generated using the DNN. This may include, for example, the electronic device 101 performing operations 210 to 213 to generate a dense depth map using the DNN. At step 1007, a disparity map between the first image and the second image is generated based on the dense depth map. This may include, for example, the electronic device 101 performing operation 410 to create a disparity map between the left corrected image 214 and the right corrected image 215 using the dense depth map.

[0114] At step 1009, a verified depth map is generated based on a pixel-level comparison of the predicted and extracted positions associated with at least some pixels in at least one of the first and second images. The predicted positions are determined based on the disparity map. This may include, for example, the electronic device 101 performing the depth clarification and verification process 230 to generate the verified depth map 218.

[0115] At step 1011, a first virtual view and a second virtual view are generated based on the verified depth map to be presented on the display panel of the VST AR device. This may include, for example, the electronic device 101 generating a final left virtual view 226 and a final right virtual view 227 for display on the left panel 926 and the right panel 927 of the VST AR head mounted device. At step 1013, the first virtual view and the second virtual view are presented on the display panel of the VST AR device. This may include, for example, the electronic device 101 displaying the final left virtual view 226 and the final right virtual view 227 on the left panel 926 and the right panel 927, respectively.

[0116] although Fig.10 An example of a method 1000 for parallax correction for video see-through augmented reality is shown, but may be used for Fig.10 For example, although shown as a series of steps, Fig.10 The various steps in may overlap, occur in parallel, occur in a different order, or occur any number of times.

[0117] As described above, the disclosed embodiments provide techniques for depth correction and noise reduction. The disclosed embodiments can generate better depth maps for depth-based image reprojection. Compared to conventional techniques (which can generate visible noise on the created virtual view), the disclosed high-quality depth maps can generate high-quality virtual camera views on the head-mounted device. The disclosed embodiments also provide improved techniques for parallax correction for different real-world camera viewpoints and generated virtual objects. The disclosed techniques can use reading world depth information to set virtual objects to virtual views with the same parallax as the real-world information from the perspective camera. The disclosed techniques can also use one or more of depth clarification, depth verification, depth correction, and noise reduction to handle incorrect depth situations, such as objects in the scene having different colors in different areas.

[0118] Although the present disclosure has been described with reference to various exemplary embodiments, various changes and modifications may be suggested to one skilled in the art. The present disclosure is intended to encompass such changes and modifications as fall within the scope of the appended claims.

Claims

1. A method comprising: Obtaining a stereoscopic image pair including a first image and a second image, wherein the first image and the second image are captured using a first perspective camera and a second perspective camera associated with a video see-through (VST) augmented reality (AR) device; generating a first feature map of the first image and a second feature map of the second image, wherein the first feature map comprises extraction locations associated with pixels in the first image and the second feature map comprises extraction locations associated with pixels in the second image; generating a disparity map between the first image and the second image based on the dense depth map; generating a verified depth map based on a pixel-level comparison of predicted and extracted positions associated with at least some pixels in at least one of the first image and the second image, wherein the predicted positions are determined based on the disparity map; and A first virtual view and a second virtual view are generated based on the verified depth map for display on a display panel of the VST AR device.

2. The method of claim 1, wherein: The steps of generating the verified depth map include: For each pixel in the first image representing a point in space: determining a predicted position of the pixel in the second image based on an extracted position of a pixel representing the point in the first image feature map and a disparity value associated with the pixel from the disparity map; comparing the predicted position of the pixel in the second image with the extracted position of the pixel in the second image feature map; and Do one of the following: In response to a difference between the predicted position and the extracted position being less than a threshold, storing in the verified depth map a depth value associated with the pixel in the dense depth map; and In response to a difference between the predicted position and the extracted position being equal to or greater than the threshold, a corrected depth value is determined and stored in the verified depth map.

3. The method of claim 2, wherein: The predicted position of the pixel in the second image is determined using the following equation: Among them, x r represents the predicted position of the pixel in the second image, x l represents the extraction position of the pixel in the first image feature map, B represents the distance between the first virtual camera position and the second virtual camera position associated with the VST AR device, f represents the focal length, and d represents the depth value associated with the pixel.

4. The method of claim 2, wherein: The corrected depth value is determined based on depth information, color texture information, and spatial information from at least one of the first image and the second image.

5. The method of claim 1, wherein: The step of generating the first virtual view and the second virtual view based on the verified depth map comprises: warping the verified depth map to generate a first warped depth map for the first virtual camera and a second warped depth map for the second virtual camera; generating a first virtual object and a second virtual object using the first warped depth map and the second warped depth map; generating the first virtual view using the first warped depth map and real-world information; and The second virtual view is generated using the second warped depth map and the real-world information.

6. The method of claim 1, further comprising: A deep neural network (DNN) is used to generate the dense depth map.

7. The method of claim 1, further comprising: The first virtual view and the second virtual view are presented on a display panel of the VST AR device.

8. A video perspective VST augmented reality (AR) device, comprising: at least one display panel; a first perspective camera and a second perspective camera; as well as At least one processing device configured to: obtaining a stereoscopic image pair including a first image and a second image captured using the first perspective camera and the second perspective camera; generating a first feature map of the first image and a second feature map of the second image, wherein the first feature map comprises extraction locations associated with pixels in the first image and the second feature map comprises extraction locations associated with pixels in the second image; generating a disparity map between the first image and the second image based on the dense depth map; generating a verified depth map based on a pixel-level comparison of predicted and extracted positions associated with at least some pixels in at least one of the first image and the second image, wherein the predicted positions are determined based on the disparity map; and A first virtual view and a second virtual view are generated based on the verified depth map for display on a display panel of the VST AR device.

9. The VST AR device according to claim 8, wherein: In order to generate the verified depth map, the at least one processing device is configured to: For each pixel in the first image representing a point in space: determining a predicted position of the pixel in the second image based on an extracted position of a pixel representing the point in the first image feature map and a disparity value associated with the pixel from the disparity map; comparing the predicted position of the pixel in the second image with the extracted position of the pixel in the second image feature map; as well as Do one of the following: In response to a difference between the predicted position and the extracted position being less than a threshold, storing a depth value associated with the pixel in the dense depth map in the verified depth map; as well as In response to a difference between the predicted position and the extracted position being equal to or greater than the threshold, a corrected depth value is determined and stored in the verified depth map.

10. The VST AR device according to claim 9, wherein: The at least one processing device is configured to determine a predicted position of the pixel in the second image using the following equation: Among them, x r represents the predicted position of the pixel in the second image, x l represents the extraction position of the pixel in the first image feature map, B represents the distance between the first virtual camera position and the second virtual camera position associated with the VST AR device, f represents the focal length, and d represents the depth value associated with the pixel.

11. The VST AR device according to claim 9, wherein: The at least one processing device is configured to determine the corrected depth value based on depth information, color texture information, and spatial information from at least one of the first image and the second image.

12. The VST AR device according to claim 8, wherein: In order to generate the first virtual view and the second virtual view based on the verified depth map, the at least one processing device is configured to: warping the verified depth map to generate a first warped depth map for the first virtual camera and a second warped depth map for the second virtual camera; generating a first virtual object and a second virtual object using the first warped depth map and the second warped depth map; generating the first virtual view using the first warped depth map and real-world information; as well as The second virtual view is generated using the second warped depth map and the real-world information.

13. The VST AR device according to claim 8, wherein: The at least one processing device is further configured to: A deep neural network (DNN) is used to generate the dense depth map.

14. The VST AR device according to claim 8, wherein: The at least one processing device is further configured to: The display panel is controlled to present the first virtual view and the second virtual view.

15. A computer readable medium comprising instructions, wherein: When the instructions are executed, the at least one processor is caused to perform the method according to any one of claims 1 to 7.

Images