Smooth alignment transition in multi-camera systems
By adjusting and aligning FOV using motion vectors and warp transformations, the method addresses FOV jumps in multi-camera systems with large baselines, improving image smoothness and user experience.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- QUALCOMM INC
- Filing Date
- 2024-11-26
- Publication Date
- 2026-06-04
AI Technical Summary
Multi-camera systems with large baselines face challenges in achieving smooth alignment transition during operations like zooming, particularly when capturing scenes with multiple depths, leading to field of view (FOV) jumps and aberrations in the output images.
A method involving a first image capture device with a larger field of view than a second device, where a motion vector is assessed to adjust and scale the FOV of the first device, followed by warp transformations and bokeh effects to align and blend foreground and background, ensuring smooth transitions.
This approach improves smooth alignment transition by mitigating FOV jumps and artifacts, enhancing user experience and efficiency in multi-camera systems.
Smart Images

Figure CN2024134438_04062026_PF_FP_ABST
Abstract
Description
SMOOTH ALIGNMENT TRANSITION IN MULTI-CAMERA SYSTEMSTECHNICAL FIELD
[0001] Aspects of the present disclosure relate generally to image processing, and more particularly, to improving smooth alignment transition in multi-camera systems. Some features may enable and provide improved image processing, including improved mitigation of field of view (FOV) jumps in images. INTRODUCTION
[0002] Image capture devices are devices that can capture one or more digital images, whether still images for photos or sequences of images for videos. Capture devices can be incorporated into a wide variety of devices. By way of example, image capture devices may comprise stand-alone digital cameras or digital video camcorders, camera-equipped wireless communication device handsets, such as mobile telephones, cellular or satellite radio telephones, personal digital assistants (PDAs) , panels or tablets, gaming devices, computing devices such as webcams, video surveillance cameras, or other devices with digital imaging or video capabilities.
[0003] In certain scenes, a user of an image capture device may desire to focus to one portion of the scene. For example, a user of a remotely controlled drone capturing images and / or videos of a sky may desire to focus on a plane flying through the sky, rather than the clouds and other scenery in the background. In this regard, multi-camera systems used by the drone and other devices may perform zooming and other functions to allow a user to focus on a region of interest. Multi-camera systems are systems employing multiple image capture devices to capture and analyze visual information in real-time. Multi-camera systems may provide numerous benefits including but not limited to increased field of view (FOV) , reduced lens distortion, and improved object tracking. Although multi-camera systems often rely on separate image data captured from disparate cameras, there is a desire for multi-camera systems to present the output image frames to a user in a manner that smoothly aligns the separately obtained image data to reduce obvious transitions or discrepancies between the image data, in a process referred to as smooth alignment transition.
[0004] However, most multi-camera systems, such as those used in drones or advanced driving assistance systems (ADAS) , are constructed with image capture devices having large baselines (e.g., at least about 3 centimeters) , the distance between two image capture devices. Large baselines often prevent or otherwise hinder the smooth alignment of image data from the multiple image capture devices in multi-camera systems. For example, it is more challenging to maintain a stable transition of image data stream from a multi-camera system with a larger baseline during zooming operations because of the difficulty in spatial alignment of image data generated by the multiple image capture devices used in the multi-camera systems.
[0005] Furthermore, multiple depths in a field of view of an image capture device of the multi-camera system often create complications when zooming and other operations are performed in the multi-camera system. Specifically, image streams from multi-camera systems often cause large shifts or jumps between the foregrounds and backgrounds of the scene being captured during such operations due to the disparity between individual image capture devices used in the multi-camera system.
[0006] Various embodiments of the present disclosure address one or more of the above described shortcomings. BRIEF SUMMARY OF SOME EXAMPLES
[0007] The following summarizes some aspects of the present disclosure to provide a basic understanding of the discussed technology. This summary is not an extensive overview of all contemplated features of the disclosure and is intended neither to identify key or critical elements of all aspects of the disclosure nor to delineate the scope of any or all aspects of the disclosure. Its sole purpose is to present some concepts of one or more aspects of the disclosure in summary form as a prelude to the more detailed description that is presented later.
[0008] There is a desire and need for multi-camera systems to have smooth alignment transition when undergoing operations such as zooming. In multi-camera systems having large baselines, smooth alignment transition is particularly difficult to achieve. Furthermore, smooth alignment transition is also difficult to achieve in multi-camera systems capturing scenes having multiple depths, as zoom operations using the multi-camera systems often result in field of view (FOV) jumps (e.g., between a background and a foreground portion) , aberrations, and undesired artifacts in the outputted images or videos, thereby affecting user experience and user decision making. The present disclosure addresses these shortcomings by ensuring smooth alignment transition in multi-camera systems, especially those with larger baselines.
[0009] In some aspects, a zooming procedure (e.g., a zoom out procedure) may be performed by a multi-camera system having at least a first image capture device and a second image capture device, where the first image capture device is configured to have a larger field of view than the field of view of the second image capture device. The zooming procedure (e.g., zoom out procedure) may occur during a switching period from the second image capture device (that has a relatively smaller FOV) to the first image capture device (that has a relatively larger FOV) to ensure improved smooth alignment transition. Furthermore, the zooming procedure may be based on a zoom request to zoom in to or zoom out from an object or other feature in an image datastream. For example, a user viewing an image datastream output from the multi-camera system maybe examining a moving object from a smaller FOV, but may desire to zoom out and view a larger FOV relative to the object (e.g., due to the object moving quickly out of the smaller FOV) . Also or alternatively, a user may seek a smooth alignment transition in a viewing experience from the smaller FOV provided by the second image capture device to the larger FOV provided by the first image capture device. In some aspects, the second image capture device, configured to capture image data (second image data) having the relatively smaller FOV, may be moved to capture the object within the center of its FOV. In some embodiments, a motion vector directed to the first image capture device from the second image capture device may be assessed. The motion vector device may indicate the distance and direction from the location of the object within the current FOV of the second image data to the center of the FOV of the image data obtained by the first image capture device (first image data) . However, as the location of the object may be at or near the center of the FOV of the second image data (e.g., based on movement of the second image capture device to capture the object at or near the center of its FOV) , the motion vector may also indicate the distance and direction from the center of the FOV of the second image data to the center of the FOV of the first image data.
[0010] If the motion vector satisfies a predetermined threshold (e.g., is sufficiently great in magnitude) , the FOV of the image data captured from the first image capture device (first image data) may be adjusted (e.g., recentered) based on the FOV center of the image captured by the second image capture device (second image data) . Furthermore, the adjusted (e.g., recentered) first image data may be scaled towards achieving the zoom ratio of the second image data. In some embodiments, the adjusting and scaling of the FOV of the first image data may be performed by way of applying a non-center crop to the first image data to generate the updated FOV of the first image data having the object at the center of the updated FOV and having the zoom ratio of the updated FOV reflect (e.g., to within a predetermined tolerance level) the zoom ratio of the FOV of the second image data. For example, a crop may be applied to the first image data, such that the crop is offset from the center of the FOV of the first image data by a distance that is based on (e.g., proportional to a magnitude of) the motion vector. Also or alternatively, the center of the FOV of the first image data may be shifted towards a center of a crop applied to the first image data such that, prior to the shifting, the center of the crop is offset from the center of the FOV of the first image data based on the motion vector.
[0011] Subsequently, the second image data may undergo one or more warp transformations (e.g., rotation warp, perspective warp, hybrid warp) to align the FOV of the second image data with the FOV of the adjusted (e.g., recentered) and scaled first image data. In some embodiments, image data (third image data) may be generated based on the alignment of the second image data with the adjusted (e.g., recentered) and scaled first image data. The third image data may be segmented to distinguish between foreground and background (e.g., using depth information) . A bokeh effect (e.g. blur) may be applied to the background to conceal or mitigate the risk of an FOV jump between the background and the foreground. In some aspects, the boundary between the foreground and the background may be further blended to provide a smoother transition between the foreground and the background. After application of the bokeh effect and blending, the modified third image data may be output (e.g., as fourth image data) by the multi-camera system.
[0012] The adjusting (e.g., recentering) of the first image data, and the warp transformation and alignment of the second image data with the first image data, allow for an improved FOV transition between image data from their respective image capture devices (e.g., first image data from the first image capture device and second image data from the second image capture device) of the multi-camera system, thus improving smooth alignment transition. Furthermore, the use of a threshold in the magnitude of the motion vector to determine whether to perform the aforementioned techniques provides for more efficient image processing and power consumption, for example, by applying the aforementioned techniques when there would otherwise be a sufficient enough risk of FOV jumps (e.g., when the motion vector is sufficiently large enough to warrant mitigating the risk) . Even further, the blurring of the background and the blending of the boundary between the background and the foreground in the third image data after the aforementioned warp transformation and alignment steps helps to conceal or otherwise mitigate the risk of FOV jumps in scenes having multiple depths (e.g., foreground and background) , thereby further improving smooth alignment transition in multi-camera systems.
[0013] In one aspect of the disclosure, a method for image processing includes receiving, from a first image sensor, first image data; receiving, from a second image sensor, second image data, wherein the first image data has a larger field of view (FOV) than the FOV of the second image data; determining, based on the first image data and the second image data, a motion vector; adjusting, based on the motion vector, the FOV of the first image data; applying a warp transformation to the second image data, wherein the warp transformation adjusts the FOV of the second image data towards the adjusted FOV of the first image data; and determining output image data based on the first image data and the second image data.
[0014] In some embodiments, adjusting the FOV of the first image data, applying the warp transformation, and determining the output image data are based on the motion vector meeting one or more criteria.
[0015] In some embodiments, adjusting the FOV of the first image data includes: shifting a center of the FOV of the first image data towards a center of a crop applied to the first image data, wherein, prior to the shifting, the center of the crop is offset from the center of the FOV of the first image data based on the motion vector.
[0016] In some embodiments, the method further includes: prior to applying the warp transformation, scaling the first image data based on a zoom ratio of the FOV of the second image data.
[0017] In some embodiments, the method further includes: prior to determining the motion vector, receiving a request to zoom out from an FOV of a previously output image data based on the FOV of the second image data; and determining the motion vector in response to the request. In some embodiments, the motion vector indicates a distance between a center of the FOV of the second image data and the center of the FOV of the first image data.
[0018] In some embodiments, applying the warp transformation includes: generating a warping matrix based on the adjusted FOV of the first image data and the FOV of the second image data; and determining a third image data by applying the warping matrix to the FOV of the second image data, wherein the output image data is further based on the third image data.
[0019] In some embodiments, determining the output image data further includes: segmenting the third image data into a plurality of portions comprising a foreground portion and a background portion using depth information associated with the third image data; applying a blur to the background portion of the third image data to generate a blurred background portion; and generating, based on the third image data, a fourth image data by blending a boundary between the foreground portion and the blurred background portion, wherein the output image data is further based on the fourth image data.
[0020] In some embodiments, segmenting the third image data includes: determining, by applying the third image data into a machine learning model, the depth information associated with the third image data.
[0021] In some embodiments, the plurality of portions further includes a boundary portion between the background portion and the foreground portion. Furthermore, generating the fourth image data may include: applying an alpha blending to the boundary portion of the third image data. After the application of the alpha blending, pixels of the boundary portion farther from the foreground portion can have a greater degree of blur than pixels of the boundary portion closer to the foreground portion, but can have a lesser degree of blur than a degree of the blur applied to the background portion.
[0022] In an additional aspect of the disclosure, an apparatus includes at least one processor and a memory coupled to the at least one processor. The at least one processor is configured to perform operations including receiving, from a first image sensor, first image data; receiving, from a second image sensor, second image data, where the first image data has a larger field of view (FOV) than the FOV of the second image data; determining, based on the first image data and the second image data, a motion vector; adjust, based on the motion vector, the FOV of the first image data; and apply a warp transformation to the second image data. The warp transformation adjusts the FOV of the second image data towards the offset FOV of the first image data. The operations may further include determining output image data based on the first image data and the second image data.
[0023] In some embodiments, the one or more processors are configured to adjust the FOV of the first image data, apply the warp transformation, and determine the output image data based on the motion vector meeting one or more criteria.
[0024] In some embodiments, the one or more processors are configured to adjust the FOV of the first image data by: shifting a center of the FOV of the first image data towards a center of a crop applied to the first image data. Prior to the shifting, the center of the crop is offset from the center of the FOV of the first image data based on the motion vector.
[0025] In some embodiments, the one or more processors are further configured to: prior to applying the warp transformation, scale the first image data based on a zoom ratio of the FOV of the second image data.
[0026] In some embodiments, the one or more processors are further configured to: prior to determining the motion vector, receive a request to zoom out from an FOV of a previously output image data based on the FOV of the second image data; and determine the motion vector in response to the request, wherein the motion vector indicates a distance between a center of the FOV of the second image data and the center of the FOV of the first image data.
[0027] In some embodiments, the one or more processors are further configured to apply the warp transformation by: generating a warping matrix based on the offset FOV of the first image data and the FOV of the second image data; and determining a third image data by applying the warping matrix to the FOV of the second image data, wherein the output image data is further based on the third image data.
[0028] In some embodiments, the one or more processors are further configured to determine the output image data by: segmenting the third image data into a plurality of portions comprising a foreground portion and a background portion using depth information associated with the third image data; applying a blur to the background portion of the third image data to generate a blurred background portion; and generating, based on the third image data, a fourth image data by blending a boundary between the foreground portion and the blurred background portion, wherein the output image data is further based on the fourth image data.
[0029] In some embodiments, the one or more processors are further configured to segment the third image data by: determining, by applying the third image data into a machine learning model, the depth information associated with the third image data.
[0030] In some embodiments, the plurality of portions further includes a boundary portion between the background portion and the foreground portion, wherein the one or more processors are further configured to generate the fourth image data by: applying an alpha blending to the boundary portion of the third image data, wherein, after the application of the alpha blending, pixels of the boundary portion farther from the foreground portion have a greater degree of blur than pixels of the boundary portion closer to the foreground portion, but have a lesser degree of blur than a degree of the blur applied to the background portion.
[0031] In an additional aspect of the disclosure, an image capture device includes: a first image sensor configured to capture a first field of view (FOV) ; a second image sensor configured to capture a second FOV, wherein the first FOV is larger than the second FOV; a memory; and one or more processors coupled to the memory. In some implementation, the the one or more processor being configured to: receive, from the first image sensor, first image data; receive, from the second image sensor, second image data; determine, based on the first image data and the second image data, a motion vector; adjust, based on the motion vector, the FOV of the first image data; apply a warp transformation to the second image data to generate a third image data, wherein the warp transformation adjusts the FOV of the second image data towards the adjusted FOV of the first image data; segment the third image data into a plurality of portions comprising a foreground portion and a background portion using depth information associated with the third image data; apply a blur to the background portion of the third image data to generate a blurred background portion; generate, based on the third image data, a fourth image data by blending a boundary between the foreground portion and the blurred background portion; and determine output image data based on the fourth image data.
[0032] In some embodiments, the plurality of portions further includes a boundary portion between the background portion and the foreground portion. The one or more processors are further configured to generate the fourth image data by: applying an alpha blending to the boundary portion of the third image data to generate the fourth image data. After application of the alpha blending, pixels of the boundary portion farther from the foreground portion have a greater degree of blur than pixels of the boundary portion closer to the foreground portion, but have a lesser degree of blur than a degree of the blur applied to the background portion.
[0033] In an additional aspect of the disclosure, an apparatus includes means for receiving, from a first image sensor, first image data; means for receiving, from a second image sensor, second image data, where the first image data has a larger field of view (FOV) than the FOV of the second image data; means for determining, based on the first image data and the second image data, a motion vector; means for adjusting, based on the motion vector, the FOV of the first image data; means for applying a warp transformation to the second image data, where the warp transformation adjusts the FOV of the second image data towards the adjusted FOV of the first image data; and means for determining output image data based on the first image data and the second image data.
[0034] In an additional aspect of the disclosure, a non-transitory computer-readable medium stores instructions that, when executed by a processor, cause the processor to perform operations. The operations include receiving, from a second image sensor, second image data, where the first image data has a larger field of view (FOV) than the FOV of the second image data; means for determining, based on the first image data and the second image data, a motion vector; means for adjusting, based on the motion vector, the FOV of the first image data; means for applying a warp transformation to the second image data, where the warp transformation adjusts the FOV of the second image data towards the adjusted FOV of the first image data; and means for determining output image data based on the first image data and the second image data.
[0035] Methods of image processing described herein may be performed by an image capture device and / or performed on image data captured by one or more image capture devices. Image capture devices, devices that can capture one or more digital images, whether still image photos or sequences of images for videos, can be incorporated into a wide variety of devices. By way of example, image capture devices may comprise stand-alone digital cameras or digital video camcorders, camera-equipped wireless communication device handsets, such as mobile telephones, cellular or satellite radio telephones, personal digital assistants (PDAs) , panels or tablets, gaming devices, computing devices such as webcams, video surveillance cameras, or other devices with digital imaging or video capabilities.
[0036] The image processing techniques described herein may involve digital cameras having image sensors and processing circuitry (e.g., application specific integrated circuits (ASICs) , digital signal processors (DSP) , graphics processing unit (GPU) , or central processing units (CPU) ) . An image signal processor (ISP) may include one or more of these processing circuits and configured to perform operations to obtain the image data for processing according to the image processing techniques described herein and / or involved in the image processing techniques described herein. The ISP may be configured to control the capture of image frames from one or more image sensors and determine one or more image frames from the one or more image sensors to generate a view of a scene in an output image frame, including a zoomed in or zoomed out view of the scene that avoids, conceals, and / or otherwise mitigates jumps in fields of view (FOV) associated with the image sensors. The output image frame may be part of a sequence of image frames forming a video sequence. The video sequence may include other image frames received from the image sensor or other images sensors.
[0037] In an example application, the image signal processor (ISP) may receive an instruction to capture a sequence of image frames in response to the loading of software, such as a camera application, to produce a preview display from the image capture device. The image signal processor may be configured to produce a single flow of output image frames, based on images frames received from one or more image sensors. The single flow of output image frames may include raw image data from an image sensor, binned image data from an image sensor, or corrected image data processed by one or more algorithms within the image signal processor. For example, an image frame obtained from an image sensor, which may have performed some processing on the data before output to the image signal processor, may be processed in the image signal processor by processing the image frame through an image post-processing engine (IPE) and / or other image processing circuitry for performing one or more of tone mapping, portrait lighting, contrast enhancement, gamma correction, etc. The output image frame from the ISP may be stored in memory and retrieved by an application processor executing the camera application, which may perform further processing on the output image frame to adjust an appearance of the output image frame and reproduce the output image frame on a display for view by the user.
[0038] After an output image frame representing the scene is determined by the image signal processor and / or determined by the application processor, such as through image processing techniques described in various embodiments herein, the output image frame may be displayed on a device display as a single still image and / or as part of a video sequence, saved to a storage device as a picture or a video sequence, transmitted over a network, and / or printed to an output medium. For example, the image signal processor (ISP) may be configured to obtain input frames of image data (e.g., pixel values) from the one or more image sensors, and in turn, produce corresponding output image frames (e.g., preview display frames, still-image captures, frames for video, frames for object tracking, etc. ) . In other examples, the image signal processor may output image frames to various output devices and / or camera modules for further processing, such as for 3A parameter synchronization (e.g., automatic focus (AF) , automatic white balance (AWB) , and automatic exposure control (AEC) ) , producing a video file via the output frames, configuring frames for display, configuring frames for storage, transmitting the frames through a network connection, etc. Generally, the image signal processor (ISP) may obtain incoming frames from one or more image sensors and produce and output a flow of output frames to various output destinations.
[0039] In some aspects, the output image frame may be produced by combining aspects of the image correction of this disclosure with other computational photography techniques such as high dynamic range (HDR) photography or multi-frame noise reduction (MFNR) . With HDR photography, a first image frame and a second image frame are captured using different exposure times, different apertures, different lenses, and / or other characteristics that may result in improved dynamic range of a fused image when the two image frames are combined. In some aspects, the method may be performed for MFNR photography in which the first image frame and a second image frame are captured using the same or different exposure times and fused to generate a corrected first image frame with reduced noise compared to the captured first image frame.
[0040] In some aspects, a device may include an image signal processor or a processor (e.g., an application processor) including specific functionality for camera controls and / or processing, such as enabling or disabling the binning module or otherwise controlling aspects of the image correction. The methods and techniques described herein may be entirely performed by the image signal processor or a processor, or various operations may be split between the image signal processor and a processor, and in some aspects split across additional processors.
[0041] The device may include one, two, or more image sensors, such as a first image sensor. When multiple image sensors are present, the image sensors may be differently configured. For example, the first image sensor may have a larger FOV than the second image sensor, or the first image sensor may have different sensitivity or different dynamic range than the second image sensor. In one example, the first image sensor may be a wide-angle image sensor, and the second image sensor may be a tele image sensor. In another example, the first sensor is configured to obtain an image through a first lens with a first optical axis and the second sensor is configured to obtain an image through a second lens with a second optical axis different from the first optical axis. Additionally or alternatively, the first lens may have a first magnification, and the second lens may have a second magnification different from the first magnification. Any of these or other configurations may be part of a lens cluster on a mobile device, such as where multiple image sensors and associated lenses are located in offset locations on a frontside or a backside of the mobile device. Additional image sensors may be included with larger, smaller, or same fields of view. The image processing techniques described herein may be applied to image frames captured from any of the image sensors in a multi-sensor device.
[0042] In an additional aspect of the disclosure, a device configured for image processing and / or image capture is disclosed. The apparatus includes means for capturing image frames. The apparatus further includes one or more means for capturing data representative of a scene, such as image sensors (including charge-coupled devices (CCDs) , Bayer-filter sensors, infrared (IR) detectors, ultraviolet (UV) detectors, complimentary metal-oxide-semiconductor (CMOS) sensors) and time of flight detectors. The apparatus may further include one or more means for accumulating and / or focusing light rays into the one or more image sensors (including simple lenses, compound lenses, spherical lenses, and non-spherical lenses) . These components may be controlled to capture the first and / or second image frames input to the image processing techniques described herein.
[0043] Other aspects, features, and implementations will become apparent to those of ordinary skill in the art, upon reviewing the following description of specific, exemplary aspects in conjunction with the accompanying figures. While features may be discussed relative to certain aspects and figures below, various aspects may include one or more of the advantageous features discussed herein. In other words, while one or more aspects may be discussed as having certain advantageous features, one or more of such features may also be used in accordance with the various aspects. In similar fashion, while exemplary aspects may be discussed below as device, system, or method aspects, the exemplary aspects may be implemented in various devices, systems, and methods.
[0044] The method may be embedded in a computer-readable medium as computer program code comprising instructions that cause a processor to perform the steps of the method. In some embodiments, the processor may be part of a mobile device including a first network adaptor configured to transmit data, such as images or videos in a recording or as streaming data, over a first network connection of a plurality of network connections; and a processor coupled to the first network adaptor and the memory. The processor may cause the transmission of output image frames described herein over a wireless communications network such as a 5G NR communication network.
[0045] The foregoing has outlined, rather broadly, the features and technical advantages of examples according to the disclosure in order that the detailed description that follows may be better understood. Additional features and advantages will be described hereinafter. The conception and specific examples disclosed may be readily utilized as a basis for modifying or designing other structures for carrying out the same purposes of the present disclosure. Such equivalent constructions do not depart from the scope of the appended claims. Characteristics of the concepts disclosed herein, both their organization and method of operation, together with associated advantages will be better understood from the following description when considered in connection with the accompanying figures. Each of the figures is provided for the purposes of illustration and description, and not as a definition of the limits of the claims.
[0046] While aspects and implementations are described in this application by illustration to some examples, those skilled in the art will understand that additional implementations and use cases may come about in many different arrangements and scenarios. Innovations described herein may be implemented across many differing platform types, devices, systems, shapes, sizes, and packaging arrangements. For example, aspects and / or uses may come about via integrated chip implementations and other non-module-component based devices (e.g., end-user devices, vehicles, communication devices, computing devices, industrial equipment, retail / purchasing devices, medical devices, artificial intelligence (AI) -enabled devices, etc. ) . While some examples may or may not be specifically directed to use cases or applications, a wide assortment of applicability of described innovations may occur. Implementations may range in spectrum from chip-level or modular components to non-modular, non-chip-level implementations and further to aggregate, distributed, or original equipment manufacturer (OEM) devices or systems incorporating one or more aspects of the described innovations. In some practical settings, devices incorporating described aspects and features may also necessarily include additional components and features for implementation and practice of claimed and described aspects. For example, transmission and reception of wireless signals necessarily includes a number of components for analog and digital purposes (e.g., hardware components including antenna, radio frequency (RF) -chains, power amplifiers, modulators, buffer, processor (s) , interleaver, adders / summers, etc. ) . It is intended that innovations described herein may be practiced in a wide variety of devices, chip-level components, systems, distributed arrangements, end-user devices, etc. of varying sizes, shapes, and constitution.BRIEF DESCRIPTION OF THE DRAWINGS
[0047] A further understanding of the nature and advantages of the present disclosure may be realized by reference to the following drawings. In the appended figures, similar components or features may have the same reference label. Further, various components of the same type may be distinguished by following the reference label by a dash and a second label that distinguishes among the similar components. If just the first reference label is used in the specification, the description is applicable to any one of the similar components having the same first reference label irrespective of the second reference label.
[0048] Figure 1 shows a block diagram of an example device for performing image capture from one or more image sensors.
[0049] Figure 2 is a block diagram illustrating an example data flow path for image data processing in an image capture device according to one or more embodiments of the disclosure.
[0050] Figure 3 shows a flow chart of an example method for processing image data from a multi-camera system for improved smooth alignment transition according to some embodiments of the disclosure.
[0051] Figure 4 shows a flow chart of an example method for processing image data of a multi-camera system for improved smooth alignment transition by using a bokeh effect to conceal or mitigate field of view (FOV) jumps in the multi-camera system according to some embodiments of the disclosure.
[0052] Figure 5 is a block diagram illustrating an example process for improved smooth alignment transition in a multi-camera system according to some embodiments of the disclosure.
[0053] Like reference numbers and designations in the various drawings indicate like elements.DETAILED DESCRIPTION
[0054] The detailed description set forth below, in connection with the appended drawings, is intended as a description of various configurations and is not intended to limit the scope of the disclosure. Rather, the detailed description includes specific details for the purpose of providing a thorough understanding of the inventive subject matter. It will be apparent to those skilled in the art that these specific details are not required in every case and that, in some instances, well-known structures and components are shown in block diagram form for clarity of presentation.
[0055] The present disclosure provides systems, apparatus, methods, and computer-readable media that support image processing, including techniques for improving smooth alignment transition in a multi-camera system, such as by mitigating field of view (FOV) jumps during a zooming process of the multi-camera system. As previously discussed, in multi-camera systems having large baselines, smooth alignment transition is particularly difficult to achieve, as the disparity in the image capture devices present issues with image alignment and cause FOV jumps. For example, for scenes having multiple depths, smooth alignment transition is difficult to achieve as zoom operations using the multi-camera systems often result in FOV jumps when viewing different depths in the outputted images or videos. In particular, objects of an outputted image or video located in greater depth (e.g., the background) have a larger FOV jump than objects of the outputted image or video located in lesser depth (e.g., the foreground) . Conventional smooth alignment transition techniques attempt to address the FOV jumps through fusion animation, or the blending of smaller FOV images captured by one image capture device of the multi-camera system onto larger FOV images captured by another image capture device of the multi-camera system. However fusion animation and other conventional smooth alignment transition techniques are known to cause various aberrations and artifacts in the outputted images or videos, such as but not limited to discontinuity between foreground and background, blurring of foreground objects, and ghost effects on the foreground objects. Such aberrations and artifacts negatively affect user visualization experience and user decision-making.
[0056] Shortcomings mentioned here are only representative and are included to highlight problems that the inventors have identified with respect to existing devices and sought to improve upon. Aspects of devices described below may address some or all of the shortcomings as well as others known in the art. Aspects of the improved devices described herein may present other benefits than, and be used in other applications than, those described above.
[0057] In some aspects, a zooming procedure (e.g., a zoom out procedure) may be performed by a multi-camera system having at least a first image capture device and a second image capture device, where the first image capture device is configured to have a larger FOV than the FOV of the second image capture device. The zooming procedure may occur during a switching period from the second image capture device (having the relatively smaller FOV) to the first image capture device (having the relatively larger FOV) to ensure improved smooth alignment transition. For example, a user may be viewing a specific moving object based on image data captured by the second image capture (second image data) . However, as the second image data has a smaller FOV and the moving object may quickly escape from the smaller FOV captured by the second image capture device, the user may desire to view the moving object by zooming out to have image data having a larger FOV (e.g., based on first image data captured by the first image capture device) . In some embodiments, a motion vector directed from the center of the relatively smaller FOV of the second image capture device to the center of a relatively larger FOV of the first image capture device may be assessed. The FOV of the second image capture device may show the moving object. If the motion vector satisfies a predetermined threshold (e.g., is sufficiently great in magnitude) , the FOV center of the image data captured from the first image capture device having the larger FOV (the first image data or the larger FOV image data) may be adjusted (e.g., by applying a non-central crop, by shifting the center of the FOV towards a center of an applied non-central crop, etc. ) based on the FOV center of the image captured by the second image capture device having the smaller FOV (the second image data or the smaller FOV image data) . Furthermore, the adjusted (e.g., recentered) first image data may be scaled towards achieving the zoom ratio of the second image data. Subsequently, the second image data may undergo one or more warp transformations (e.g., rotation warp, perspective warp, hybrid warp) to align the FOV of the second image data with the FOV of the adjusted (e.g., recentered) and scaled up first image data. In some embodiments, image data (third image data) may be generated based on the alignment of the second image data with the adjusted and scaled up first image data.
[0058] In some aspects, the third image data may be segmented to distinguish between foreground and background (e.g., using depth information associated with the third image data) . A bokeh effect (e.g. blur) may be applied to the background to conceal or mitigate the risk of an FOV jump between the background and the foreground. In some aspects, the boundary between the foreground and the background may be further blended to ensure a smoother transition between the foreground and the background. After application of the bokeh effect and blending, the modified third image data may be output (e.g., as fourth image data) by the multi-camera system.
[0059] Particular implementations of the subject matter described in this disclosure may be implemented to realize one or more of the following potential advantages or benefits. In some aspects, the present disclosure provides techniques for improved smooth alignment transition in multi-camera systems, especially those with large baselines. For example, the adjusting (e.g., recentering) of the larger FOV image data the warp transformation of the smaller FOV image data based on the larger FOV image data, and the alignment of the smaller FOV image data with the larger FOV image data allow for an improved FOV transition between the image capture devices (e.g., the first image capture device and the second image capture device) of the multi-camera system, thus improving smooth alignment transition. Furthermore, the use of a threshold in the magnitude of the motion vector to determine whether to perform the aforementioned techniques provides for more efficient image processing and power consumption, for example, by applying the aforementioned techniques when there would otherwise be a risk of FOV jumps (e.g., when the motion vector is sufficiently large enough to warrant mitigating the risk) . Even further, the blurring of the background and the blending of the boundary between the background and the foreground in the third image data after the aforementioned warp transformation and alignment steps help to conceal or otherwise mitigate FOV jumps in scenes having one or more depths (e.g., foreground and background) , thereby further improving smooth alignment transition in multi-camera systems.
[0060] In the description of embodiments herein, numerous specific details are set forth, such as examples of specific components, circuits, and processes to provide a thorough understanding of the present disclosure. The term “coupled” as used herein means connected directly to or connected through one or more intervening components or circuits. Also, in the following description and for purposes of explanation, specific nomenclature is set forth to provide a thorough understanding of the present disclosure. However, it will be apparent to one skilled in the art that these specific details may not be required to practice the teachings disclosed herein. In other instances, well known circuits and devices are shown in block diagram form to avoid obscuring teachings of the present disclosure.
[0061] Some portions of the detailed descriptions which follow are presented in terms of procedures, logic blocks, processing, and other symbolic representations of operations on data bits within a computer memory. In the present disclosure, a procedure, logic block, process, or the like, is conceived to be a self-consistent sequence of steps or instructions leading to a desired result. The steps are those requiring physical manipulations of physical quantities. Usually, although not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated in a computer system.
[0062] An example device for capturing image frames using one or more image sensors, such as a smartphone, may include a configuration of one, two, three, four, or more camera modules on a backside (e.g., a side opposite a primary user display) and / or a front side (e.g., a same side as a primary user display) of the device. The devices may include one or more image signal processors (ISPs) , Computer Vision Processors (CVPs) (e.g., AI engines) , or other suitable circuitry for processing images captured by the image sensors. The one or more image signal processors (ISP) may store output image frames (such as through a bus) in a memory and / or provide the output image frames to processing circuitry (such as an applications processor) . The processing circuitry may perform further processing, such as for encoding, storage, transmission, or other manipulation of the output image frames.
[0063] As used herein, a camera module may include the image sensor and certain other components coupled to the image sensor used to obtain a representation of a scene in image data comprising an image frame. For example, a camera module may include other components of a camera, including a shutter, buffer, or other readout circuitry for accessing individual pixels of an image sensor. In some embodiments, the camera module may include one or more components including the image sensor included in a single package with an interface configured to couple the camera module to an image signal processor or other processor through a bus.
[0064] Figure 1 shows a block diagram of a device 100 for performing image capture from one or more image sensors. The device 100 may include, or otherwise be coupled to, an image signal processor (e.g., ISP 112) for processing image frames from one or more image sensors, such as a first image sensor 101, a second image sensor 102, and a depth sensor 145. In some implementations, the device 100 also includes or is coupled to a processor 104 and a memory 106 storing instructions 108 (e.g., a memory storing processor-readable code or a non-transitory computer-readable medium storing instructions) . The device 100 may also include or be coupled to a display 114 and components 116. Components 116 may be used for interacting with a user, such as a touch screen interface and / or physical buttons.
[0065] Components 116 may also include network interfaces for communicating with other devices, including a wide area network (WAN) adaptor (e.g., WAN adaptor 116a) , a local area network (LAN) adaptor (e.g., LAN adaptor 116b) , and / or a personal area network (PAN) adaptor (e.g., PAN adaptor 116c) . A WAN adaptor 116a may be a 4G LTE or a 5G NR wireless network adaptor. A LAN adaptor 116b may be an IEEE 802.11 WiFi wireless network adapter. A PAN adaptor 116c may be a Bluetooth wireless network adaptor. Each of the WAN adaptor 116a, LAN adaptor 116b, and / or PAN adaptor 116c may be coupled to an antenna, including multiple antennas configured for primary and diversity reception and / or configured for receiving specific frequency bands. In some embodiments, antennas may be shared for communicating on different networks by the WAN adaptor 116a, LAN adaptor 116b, and / or PAN adaptor 116c. In some embodiments, the WAN adaptor 116a, LAN adaptor 116b, and / or PAN adaptor 116c may share circuitry and / or be packaged together, such as when the LAN adaptor 116b and the PAN adaptor 116c are packaged as a single integrated circuit (IC) .
[0066] The device 100 may further include or be coupled to a power supply 118 for the device 100, such as a battery or an adaptor to couple the device 100 to an energy source. The device 100 may also include or be coupled to additional features or components that are not shown in Figure 1. In one example, a wireless interface, which may include a number of transceivers and a baseband processor in a radio frequency front end (RFFE) , may be coupled to or included in WAN adaptor 116a for a wireless communication device. In a further example, an analog front end (AFE) to convert analog image data to digital image data may be coupled between the first image sensor 101 or second image sensor 102 and processing circuitry in the device 100. In some embodiments, AFEs may be embedded in the ISP 112.
[0067] The device may include or be coupled to a sensor hub 150 for interfacing with sensors to receive data regarding movement of the device 100, data regarding an environment around the device 100, data regarding motion of a FOV of the one or more image sensors, data regarding depth within a FOV (e.g., obtained via the depth sensor 145) , and / or other non-camera sensor data. One example non-camera sensor is a gyroscope, which is a device configured for measuring rotation, orientation, and / or angular velocity to generate motion data. Another example non-camera sensor is an accelerometer, which is a device configured for measuring acceleration, which may also be used to determine velocity and distance traveled by appropriately integrating the measured acceleration. In some aspects, a gyroscope in an electronic image stabilization system (EIS) may be coupled to the sensor hub. In another example, a non-camera sensor may be a global positioning system (GPS) receiver, which is a device for processing satellite signals, such as through triangulation and other techniques, to determine a location of the device 100. The location may be tracked over time to determine additional motion information, such as velocity and acceleration. The data from one or more sensors may be accumulated as motion data by the sensor hub 150. One or more of the acceleration, velocity, and / or distance may be included in motion data provided by the sensor hub 150 to other components of the device 100, including the ISP 112 and / or the processor 104. In some embodiments, the sensor hub 150 may include a motion detection engine 151, which may include application-specific circuitry, software, and / or firmware configured to detect, based on the motion data obtained via the sensors, a motion vector 152. The motion vector 152 may include a magnitude of the motion and a direction of the motion. In some aspects, the motion vector 152 may indicate a displacement in a FOV. In some embodiments, the sensor hub 150 may determine or store depth information 153 based on data obtained view the depth sensor 145. In at least one embodiment, the depth information may include a depth, or a relative depth associated with one or more pixels of an image data obtained via an image sensor. In some embodiments, the depth information 153 associated with the image data may be obtained via object recognition techniques and models (e.g., trained and / or applied using AI engine 124) .
[0068] The ISP 112 may receive captured image data. In one embodiment, a local bus connection couples the ISP 112 to the first image sensor 101 and second image sensor 102 of a first camera 103 and second camera 105, respectively. In another embodiment, a wire interface couples the ISP 112 to an external image sensor. In a further embodiment, a wireless interface couples the ISP 112 to the first image sensor 101 or second image sensor 102.
[0069] The first image sensor 101 and the second image sensor 102 are configured to capture image data representing a scene in the FOV of the first camera 103 and second camera 105, respectively. In some embodiments, the first camera 103 and / or second camera 105 output analog data, which is converted by an analog front end (AFE) and / or an analog-to-digital converter (ADC) in the device 100 or embedded in the ISP 112. In some embodiments, the first camera 103 and / or second camera 105 output digital data. The digital image data may be formatted as one or more image frames, whether received from the first camera 103 and / or second camera 105 or converted from analog data received from the first camera 103 and / or second camera 105.
[0070] The first camera 103 may include the first image sensor 101 and a first lens 131. The second camera may include the second image sensor 102 and a second lens 132. Each of the first lens 131 and the second lens 132 may be controlled by an associated autofocus (AF) algorithm (e.g., AF 133) executing in the ISP 112, which adjusts the first lens 131 and the second lens 132 to focus on a particular focal plane located at a certain scene depth. The AF 133 may be assisted by depth data received from depth sensor 145. The first lens 131 and the second lens 132 focus light at the first image sensor 101 and second image sensor 102, respectively, through one or more apertures for receiving light, one or more shutters for blocking light when outside an exposure window, and / or one or more color filter arrays (CFAs) for filtering light outside of specific frequency ranges. The first lens 131 and second lens 132 may have different fields of view (FOVs) to capture different representations of a scene. In some embodiments, the first lens 131 may enable the first image sensor 101 to have a relatively larger FOV than the FOV of the second image sensor 102. Also or alternatively, the second lens 132 may enable second image sensor 102 to have a relatively smaller FOV than the FOV of the first image sensor 101. For example, the first lens 131 may be an ultra-wide (UW) lens and the second lens 132 may be a wide (W) lens. The multiple image sensors may include a combination of UW, W, tele (T) , and ultra-tele (UT) sensors.
[0071] Each of the first camera 103 and second camera 105 may be configured through hardware configuration and / or software settings to obtain different, but overlapping, FOVs. In some configurations, the cameras are configured with different lenses with different magnification ratios that result in different fields of view for capturing different representations of the scene. The cameras may be configured such that a UW camera has a larger FOV than a W camera, which has a larger FOV than a T camera, which has a larger FOV than a UT camera. For example, a camera configured for wide FOV may capture fields of view in the range of 64-84 degrees, a camera configured for ultra-side FOV may capture fields of view in the range of 100-140 degrees, a camera configured for tele FOV may capture fields of view in the range of 10-30 degrees, and a camera configured for ultra-tele FOV may capture fields of view in the range of 1-8 degrees.
[0072] In some embodiments, one or more of the first camera 103 and / or second camera 105 may be a variable aperture (VA) camera in which the aperture can be adjusted to set a particular aperture size. Example aperture sizes include f / 2.0, f / 2.8, f / 3.2, f / 8.0, etc. Larger aperture values correspond to smaller aperture sizes, and smaller aperture values correspond to larger aperture sizes. A variable aperture (VA) camera may have different characteristics that produced different representations of a scene based on a current aperture size. For example, a VA camera may capture image data with a depth of focus (DOF) corresponding to a current aperture size set for the VA camera.
[0073] The ISP 112 processes image frames captured by the first camera 103 and second camera 105. While Figure 1 illustrates the device 100 as including first camera 103 and second camera 105, any number (e.g., one, two, three, four, five, six, etc. ) of cameras may be coupled to the ISP 112. In some aspects, depth sensors such as depth sensor 145 may be coupled to the ISP 112. Output from the depth sensor 145 may be processed in a similar manner to that of first camera 103 and second camera 105. Examples of depth sensor 145 include active sensors, including one or more of indirect Time of Flight (iToF) , direct Time of Flight (dToF) , light detection and ranging (Lidar) , mmWave, radio detection and ranging (Radar) , and / or hybrid depth sensors, such as structured light sensors. In embodiments without a depth sensor 145, similar information regarding depth of objects or a depth map may be determined from the disparity between first camera 103 and second camera 105, such as by using a depth-from-disparity algorithm, a depth-from-stereo algorithm, phase detection auto-focus (PDAF) sensors, or the like. In addition, any number of additional image sensors or image signal processors may exist for the device 100.
[0074] In some embodiments, the ISP 112 may execute instructions from a memory, such as instructions 108 from the memory 106, instructions stored in a separate memory coupled to or included in the ISP 112, or instructions provided by the processor 104. In addition, or in the alternative, the ISP 112 may include specific hardware (such as one or more integrated circuits (ICs) ) configured to perform one or more operations described in the present disclosure. For example, the ISP 112 may include image front ends (e.g., IFE 135) , image post-processing engines (e.g., IPE 136) , auto exposure compensation (AEC) engines (e.g., AEC 134) , one or more engines for video analytics (e.g., EVA 137) , a warping engine 138, a segmentation engine 139, a blurring engine 140, and / or a blending engine 141. An image pipeline may be formed by a sequence of one or more of the IFE 135, IPE 136, EVA 137, warping engine 138, segmentation engine 139, blurring engine 140, and / or blending engine 141. In some embodiments, the image pipeline may be reconfigurable in the ISP 112 by changing connections between the IFE 135, IPE 136, EVA 137, warping engine 138, segmentation engine 139, blurring engine 140, and / or blending engine 141. The AF 133, AEC 134, IFE 135, IPE 136, EVA 137, warping engine 138, segmentation engine 139, blurring engine 140, and blending engine 141 may each include application-specific circuitry, be embodied as software or firmware executed by the ISP 112, and / or a combination of hardware and software or firmware executing on the ISP 112.
[0075] For example, the ISP 112 may use the warping engine 138 to apply a warp transformation to an image data obtained from one image sensor based on another image data obtained from another image sensor. For example, one or more pixels of the former image data may be adjusted towards aligning or otherwise becoming more similar to a corresponding pixel of the latter image data. In some aspects, the warp transformation may be performed via a warping matrix mapping pixels of the former image data to the latter image data. Examples of warp transformations may include but are not limited to a rotation warp, a perspective warp, or a hybrid warp.
[0076] The ISP 112 may use the segmentation engine 139 to segment a given image data into segments or portions. The segmentation may be based on a property of one or more pixels of the given image data, such as depth associated with the one or more pixels (e.g., obtained via the sensor hub 150 and depth sensor 145) . For example, the segmentation engine 139 may be used to segment a given image data into a foreground portion and a background portion based on depth information associated with various pixels of the image data, such as where pixels associated with greater depth are determined to be the background portion and pixels associated with lesser depth are determined to be the foreground portion. In some embodiments, intermediate depth may be determined as being a boundary portion between the background portion and the foreground portion. Also or alternatively, varying levels of depth may be associated with varying levels between the foreground and background portions of the image data. In some embodiments, the depth information 153 associated with the image data may be obtained via object recognition techniques and models (e.g., trained and / or applied using AI engine 124) .
[0077] The ISP 112 may use the blurring engine 140 to apply a blur to one or more pixels or portions of an image data. In some embodiments, the degree or magnitude of blur applied to a given pixel or portion of an image may depend on or may otherwise correlate with a depth associated with the given pixel or portion. For example, the blur may be applied to a pixel, a portion, or an image, by convolving the pixel, portion, or image with a kernel that may depend on the depth or distance. In some embodiments, the applied blur may be a bokeh effect, a defocus blur, and / or Gaussian blur.
[0078] The ISP 112 may use the blending engine 141 to smoothen boundaries of different segments or portions of the image, for example the foreground and background portion of the image. In some embodiments, the blending engine 141 may be configured to apply varying degrees of blur to a boundary region between the foreground portion and the background portion of an image. For example, pixels of the smoothened boundary farther from the foreground portion may be rendered to have a greater degree of blur than pixels of the smoothened boundary closer to the foreground portion, but may be rendered to have a lesser degree of blur than a degree of the blur applied to the background portion. The blending engine 141 may perform such blending by assigning or relying on assigned weights to pixels of the boundary portion, where each weight is associated with a proximity to the background portion. The blending engine 141 may apply a blur to each pixel of the boundary portion, such that a degree of blur applied to a given pixel is based on the respective weight assigned to the given pixel.
[0079] The memory 106 may include a non-transient or non-transitory computer readable medium storing computer-executable instructions as instructions 108 to perform all or a portion of one or more operations described in this disclosure. The instructions 108 may include a camera application (or other suitable application such as a messaging application) to be executed by the device 100 for photography or videography. The instructions 108 may also include other applications or programs executed by the device 100, such as an operating system and applications other than for image or video generation. Execution of the camera application, such as by the processor 104, may cause the device 100 to record images using the first camera 103 and / or second camera 105 and the ISP 112.
[0080] In addition to instructions 108, the memory 106 may also store image frames. The image frames may be output image frames stored by the ISP 112. The output image frames may be accessed by the processor 104 for further operations. In some embodiments, the device 100 does not include the memory 106. For example, the device 100 may be a circuit including the ISP 112, and the memory may be outside the device 100. The device 100 may be coupled to an external memory and configured to access the memory for writing output image frames for display or long-term storage. In some embodiments, the device 100 is a system-on-chip (SoC) that incorporates the ISP 112, the processor 104, the sensor hub 150, the memory 106, and / or components 116 into a single package.
[0081] In some embodiments, at least one of the ISP 112 or the processor 104 executes instructions to perform various operations described herein, including improving smooth alignment transition in multi-camera systems, such as by mitigating FOV jumps during zooming processes. For example, execution of the instructions can instruct the ISP 112 to begin or end capturing an image frame or a sequence of image frames, in which the capture includes correction as described in embodiments herein. In some embodiments, the processor 104 may include one or more general-purpose processor cores 104A-N capable of executing instructions to control operation of the ISP 112. For example, the cores 104A-N may execute a camera application (or other suitable application for generating images or video) stored in the memory 106 that activate or deactivate the ISP 112 for capturing image frames and / or control the ISP 112 in the application of smooth alignment transition to the image frames. The operations of the cores 104A-N and ISP 112 may be based on user input. For example, a camera application executing on processor 104 may receive a user command to begin a video preview display upon which a video comprising a sequence of image frames is captured and processed from the first camera 103 and / or the second camera 105 through the ISP 112 for display and / or storage. In some embodiments, the user may choose to zoom in or zoom out on a specific feature of a FOV captured of the output image. As the image capture device may comprise a multi-camera system based on the use of a larger FOV of the first camera 103 and a smaller FOV of the second camera 105, the zooming procedure may involve one or more techniques described herein to ensure smooth alignment transition, for example, to conceal or otherwise mitigate jumps or shifts associated with the separate FOVs of the respective cameras during the zooming process. Image processing to determine “output” or “corrected” image frames to achieve improved smooth alignment transition, such as according to techniques described herein, may be applied to one or more image frames in the sequence.
[0082] In some embodiments, the processor 104 may include ICs or other hardware (e.g., an artificial intelligence (AI) engine such as AI engine 124 or other co-processor) to offload certain tasks from the cores 104A-N. The AI engine 124 may be used to offload tasks related to, for example, depth detection, face detection, and / or object recognition performed using machine learning (ML) or artificial intelligence (AI) . The AI engine 124 may be referred to as an Artificial Intelligence Processing Unit (AI PU) . The AI engine 124 may include hardware configured to perform and accelerate convolution operations involved in executing machine learning algorithms, such as by executing predictive models such as artificial neural networks (ANNs) (including multilayer feedforward neural networks (MLFFNN) , the recurrent neural networks (RNN) , and / or the radial basis functions (RBF) ) . The ANN executed by the AI engine 124 may access predefined training weights for performing operations on user data. The ANN may alternatively be trained during operation of the image capture device 100, such as through reinforcement training, supervised training, and / or unsupervised training. In some other embodiments, the device 100 does not include the processor 104, such as when all of the described functionality is configured in the ISP 112.
[0083] In some embodiments, the display 114 may include one or more suitable displays or screens allowing for user interaction and / or to present items to the user, such as a preview of the output of the first camera 103 and / or second camera 105. In some embodiments, the display 114 is a touch-sensitive display. The input / output (I / O) components, such as components 116, may be or include any suitable mechanism, interface, or device to receive input (such as commands) from the user and to provide output to the user through the display 114. For example, the components 116 may include (but are not limited to) a graphical user interface (GUI) , a keyboard, a mouse, a microphone, speakers, a squeezable bezel, one or more buttons (such as a power button) , a slider, a toggle, or a switch.
[0084] While shown to be coupled to each other via the processor 104, components (such as the processor 104, the memory 106, the ISP 112, the display 114, and the components 116) may be coupled to each another in other various arrangements, such as via one or more local buses, which are not shown for simplicity. One example of a bus for interconnecting the components is a peripheral component interface (PCI) express (PCIe) bus.
[0085] While the ISP 112 is illustrated as separate from the processor 104, the ISP 112 may be a core of a processor 104 that is an application processor unit (APU) , included in a system on chip (SoC) , or otherwise included with the processor 104. While the device 100 is referred to in the examples herein for performing aspects of the present disclosure, some device components may not be shown in Figure 1 to prevent obscuring aspects of the present disclosure. Additionally, other components, numbers of components, or combinations of components may be included in a suitable device for performing aspects of the present disclosure. As such, the present disclosure is not limited to a specific device or configuration of components, including the device 100.
[0086] The exemplary image capture device of Figure 1 may be operated to obtain improved smooth alignment transition in multi-camera systems, such as by concealing or otherwise mitigating FOV jumps during zooming processes. One example method of operating cameras of a multi-camera system, such as first camera 103 and second camera 105, is shown in Figure 2 and described below.
[0087] Figure 2 is a block diagram illustrating an example data flow path for image data processing in an image capture device according to one or more embodiments of the disclosures. Processor 104 of system 200 may communicate with ISP 112 through a bi-directional bus and / or separate control and data lines. The processor 104 may control the first camera 103 and / or the second camera 105 through one or more camera control (s) 210. The camera control (s) 210 may be a camera driver executed by the processor 104 for configuring the first camera 103 and / or the second camera 105, such as to activate or deactivate image capture, perform zooming operations 206 (e.g., zoom in, zoom out, etc. ) , perform FOV adjustment (e.g., center offsetting, recentering, etc. ) operations 208 (e.g., based on a motion vector formed by a location in the image data that is a focus of the zooming operations) , configure exposure settings, and / or configure aperture size. Camera control (s) 210 may be managed by a camera application 204 executing on the processor 104. The camera application 204 provides settings accessible to a user such that a user can specify individual camera settings or select a profile with corresponding camera settings. Camera control (s) 210 communicates with the first camera 103 and the second camera 105 to configure the first camera 103 and the second camera 105, respectively, in accordance with commands received from the camera application 204. The camera application 204 may be, for example, a photography application, a document scanning application, a messaging application, or other application that processes image data acquired from the first camera 103.
[0088] The camera configuration may include parameters that specify, for example, a frame rate, an image resolution, a readout duration, an exposure level, an aspect ratio, an aperture size, etc. The first camera 103 and / or the second camera 105 may apply the camera configuration and obtain image data representing a scene using the camera configuration. In some embodiments, the camera configuration may be adjusted to obtain different representations of the scene. For example, the processor 104 may execute a camera application 204 to instruct the first camera 103, through camera control 210, to set a first camera configuration for the first camera 103, to obtain first image data from the first camera 103 operating in the first camera configuration, to instruct the first camera 103 to set a different camera configuration for the first camera 103, and to obtain a subsequent image data from the first camera 103 operating in the different camera configuration.
[0089] In some embodiments, the camera configuration may be adjusted to obtain a zoomed in or zoomed out representation of the scene. The zoom-in or zoom-out may occur relative to an object or point in a FOV of an image data. For example, the first image data produced by the first camera 103, which may have a larger FOV than that of the second image data produced by the second camera 105, may show an object that is within the FOV of the first image data that a user would like to further examine. The user may provide an input to zoom in to the object. In order to obtain a zoomed in representation, the processor 104 may execute a camera application 204 to instruct the first camera 103 and / or the second camera 105, through camera control (s) 210, to set a camera configuration for the first camera 103 and / or the second camera 105 to obtain the zoomed in representation according to the techniques presented herein. As another example, a user may wish to zoom out to examine a larger FOV with the object at the center. The user may provide an input to zoom out, relative to the object. In order to obtain a zoomed-out representation, the processor 104 may execute a camera application 204 to instruct the first camera 103 and / or the second camera 105, through camera control (s) 210, to set a camera configuration for the first camera 103 and / or the second camera 105 to obtain the zoomed out representation according to the techniques presented herein. The techniques for performing the aforementioned zooming operations, which may utilize ISP 112 capabilities, such as but not limited to the warping engine 138, segmentation engine 139, blurring engine 140, and blending engine 141, may deliver improved smooth alignment transition in multi-camera systems, such as by mitigating FOV jumps during the zooming processes.
[0090] Furthermore, in some embodiments, the object on which the zoom-in or zoom-out is performed may be located in a non-central location of the larger FOV of the first image data. However, the object may be located within the center of the smaller FOV of the second image data obtained by the second camera 105. In order to obtain a zoomed in representation or zoomed out representation with the object being at the center of the FOV of the outputted image data, the processor 104 may execute a camera application 204 to instruct the first camera 103, through camera control 210, to set a camera configurations for the first camera 103 to adjust a center of (e.g., recenter) the first image data from the first camera 103 so that the object is at the center of the FOV of the first image data. The adjusting may be based on a motion vector indicating the distance and direction of the object relative to the current center of the FOV of the first image data. The instructions to the first camera 103 may thus cause the FOV of the first image data to adjust based on the motion vector so that the object becomes at the center of the FOV.
[0091] In some embodiments in which the first camera 103 and / or the second image camera 105 are a part of a variable aperture (VA) camera system, the processor 104 may execute a camera application 204 to instruct the camera (s) (e.g., first camera 103 and / or second camera 105) to configure to a first aperture size, obtain image data from the camera (s) , instruct the camera (s) to configure to a second aperture size, and obtain a subsequent image data from the camera (s) . The reconfiguration of the aperture and obtaining of the original and subsequent image data may occur with little or no change in the scene captured at the first aperture size and the second aperture size. Example aperture sizes are f / 2.0, f / 2.8, f / 3.2, f / 8.0, etc. Larger aperture values correspond to smaller aperture sizes, and smaller aperture values correspond to larger aperture sizes. That is, f / 2.0 corresponds to a larger aperture size than f / 8.0.
[0092] The image data received from the first camera 103 and / or the second camera 105 may be processed in one or more blocks of the ISP 112 to determine output image frames 230 that may be stored in memory 106 and / or otherwise provided to the processor 104. The processor 104 may further process the image data to apply effects to the output image frames 230. Effects may include Bokeh, lighting, color casting, and / or high dynamic range (HDR) merging. In some embodiments, the effects may be applied in the ISP 112.
[0093] The output image frames 230 by the ISP 112 may include representations of the scene improved by aspects of this disclosure, such that the output image frames display an improvement in smooth alignment transition based on image data obtained from the multiple cameras (first camera 103 and second camera 105) of a multi-camera system 200, especially where there is a large baseline (e.g., at least about 3 cm) between the cameras (e.g., between the first camera 103 and second camera 105) . The improvement in the smooth alignment transition can manifest itself in the output image frames, for example, by a reduction of, concealment of, or otherwise a mitigation of FOV jumps in the output image frames. FOV jumps may typically occur during a zooming process and / or in areas of an image data corresponding to different depths (e.g., foreground, background, etc. ) or sources. The processor 104 may display these output image frames 230 to a user, and the improvements provided by the described processing implemented in the ISP 112 and / or processor 104 improve the image quality and the user experience by reducing, concealing, or otherwise mitigating the FOV jumps, for example, between foreground and background, reducing undesired artifacts, and making operations such as zooming be smoothly displayed in the output image frames.
[0094] Various aspects and / or components in the ISP 112 may correct the image data received from the first camera 103 and the second camera 105 of a multi-camera system 200 when determining the output image frames 230, where the first camera 103 is configured to obtain image data (first image data) having a relatively larger FOV than the FOV of the image data (second image data) obtained by the second camera 105. For example, a user may desire to zoom in or zoom out relative to an object seen in an outputted image frame, and may input the zoom request accordingly (e.g., via camera app 204) . The first camera 103 and second camera 105 may be instructed, through the zooming functionality 206 of the camera control (s) 210, to perform the zoom request, based on methodologies utilizing various aspects and / or components of the ISP 112 to provide an improved smooth alignment transition.
[0095] In at least one embodiment, the processor 104 may determine and / or store a motion vector 152 based on the zoom request. For example, the motion vector 152 may be based on whether the larger FOV of the first image data (e.g., obtained by the first camera 103) is centered at the object in which the zoom-in or zoom-out is being performed. In at least one embodiment, if the FOV of the first image data is not centered at the object, the motion vector 152 may represent the direction and distance between the center of the FOV of the first image data and the object. Also or alternatively, in some embodiments, the smaller FOV of the second image data (e.g., obtained by the second camera 105) may be centered at the object on which the zooming in to or zooming out from is being requested. In such embodiments, the motion vector 152 may represent the direction and distance from the center of the larger FOV of the first image data to the center of the FOV of the second image data (that is co-located or otherwise overlaps with the object) .
[0096] In some embodiments, performance of one or more subsequent steps to provide the improved smooth alignment transition in the outputted image frames (delivering the zoomed in or zoomed out image representations) may be predicated on the motion vector 152 satisfying one or more criteria, such as for example, a predetermined threshold distance and / or a predetermined threshold change in direction.
[0097] Based on the determined motion vector, the processor 104 may instruct the first camera 103, through camera control 210 (e.g., via a recentering functionality 208 of the camera control (s) 210) , to set a camera configuration for the first camera 103 to adjust a center of (e.g., recenter) the first image data from the first camera 103 so that the object becomes at the center of the FOV of the first image data. In some embodiments the offsetting may be performed by applying a non-center crop to the first image data, to shift the cropped center based on the motion vector towards the object. In some embodiments, a crop may be applied to the first image data, such that the crop is offset from the center of the FOV of the first image data by a distance that is based on (e.g., proportional to a magnitude of) the motion vector. Also or alternatively, the center of the FOV of the first image data may be shifted towards a center of a crop applied to the first image data such that, prior to the shifting, the center of the crop is offset from the center of the FOV of the first image data based on (e.g., proportionate to a magnitude of) the motion vector.
[0098] In some embodiments, the adjusted FOV of the first image data and / or the FOV of the second image data may both be scaled to be within the desired zoom ratio. For example, the processor 104 may instruct the first camera 103 (e.g., via the zooming functionality 206 of the camera control (s) 210) , to scale the adjusted first image data to the zoom ratio desired by the user. In some embodiments, the second image data obtained from the second camera 105 may already be at the zoom ratio desired by the user, and thus the adjusted first image data may be scaled to match or achieve a similarity threshold with the zoom ratio of the second image data. For example, the FOV of the second image data may show at its center the object that is desired to be examined (e.g., zoomed in) or zoomed out from. Alternatively, the processor 104 may instruct the second camera 105 (e.g., via the zooming functionality 206 of the camera control (s) 210) , to also scale the second image data to the zoom ratio desired by the user.
[0099] In some embodiments, the processor 104 may rely on the ISP 112 to apply a warp transformation (e.g., via warping engine 138) on the second image data to adjust the FOV of the second image data towards the adjusted FOV of the first image data. For example, pixels of the FOV of the second image data may be adjusted (e.g., via rotation, perspective changes, or a combination thereof) to align, match, or otherwise become more similar to corresponding pixels of the adjusted FOV of the first image data. In some embodiments, the warp transformation may be performed via a warping matrix configured to map pixels of the FOV of the second image data with corresponding pixels of the FOV of the first image data.
[0100] The processor 104 may rely on the ISP 112 to generate a third image data based on the fusion and / or alignment of the first image data and the second image data after one or more of the aforementioned modifications (e.g., FOV adjustments, scaling, warp transformations, etc. ) .
[0101] The improvements in smooth alignment transition delivered by techniques described herein may also include the improved focus and / or clarity of relevant aspects of an outputted image frame in a manner that avoids FOV jumps between those relevant aspects and the background of the outputted image. Such relevant aspects may be determined based on segmentation. For example, in some embodiments, the processor 104 may rely on the ISP 112 to segment (e.g., via the segmentation engine 139) the third image data into various portions, such as but not limited to the background portion 241 and a foreground portion 242. In at least one embodiment, the ISP 112 may further segment (e.g., via the segmentation engine 139) into a boundary portion situated between the background portion 241 and the foreground portion 242. In some embodiments, the segmentations may result in additional portions of varying depth or importance between the background portion 241 and the foreground portion 242.
[0102] The segmented portions may be based on depth information 153 and / or the object on which the zooming operation is performed. For example, it is contemplated that the scene captured by the FOVs of the cameras 103 and 105 may have varying levels of depth. Depth information 153 may be obtained (e.g., via depth sensor 145) from image frames 230 (e.g., of the third image data and stored. Various portions (e.g., background portion 241, foreground portion 242, and boundary portion 243) may be determined based on the depth information 153 (e.g., via the segmentation engine 139 of the ISP 112) . Also or alternatively, the zooming desired by the user may be driven by the desire to focus an outputted image on an object in contrast to the rest of the image. Thus, in some embodiments, the foreground portion 242 and background portion 242 may be based on the object and the rest of the image, respectively.
[0103] In some embodiments, the segmentation engine 139 may rely on AI models 240 trained to determine depth information 153 based on image data, and / or trained to determine segments (e.g., foreground portion 242) based on the depth information 153 and / or the underlying image data. For example, the AI models 240 may be trained based on a training dataset that includes reference image data that may be labeled to indicate different segments or portions or different depths. In some embodiments, each pixel of the image data may be assigned a weight that is indicative of a depth. The training dataset may be relied on for supervised learning to generate and / or improve the AI models 240.
[0104] The processor 104 may rely on the ISP 112 to apply a blur (e.g., via the blurring engine 140) to at least the background portion 243 of the segmented third image data. For example, a bokeh effect (e.g. blur) may be applied to the background portion 241 to conceal or mitigate the risk of an FOV jump between the background portion 241 and the foreground portion 242. In some aspects, a boundary between the foreground and the background may be further blended (e.g., via the blending engine 141) to ensure a smoother transition between the foreground portion 242 and the background portion 241. In some aspects, the boundary may comprise a separately segmented portion 243, which is subsequently blended. In at least one embodiment, the blending engine 141 may map pixels of the boundary or boundary portion based on their proximity to the background portion and / or their distance from the foreground portion. In some aspects, weights may be assigned to each pixel based on or indicative of this proximity or distance. The blending may involve applying a degree of blur to pixels of the boundary or boundary portion 243 based on the weight. For example, pixels of the smoothened boundary farther from the foreground portion 242 may have a greater degree of blur than pixels of the smoothened boundary closer to the foreground portion 242, but may have a lesser degree of blur than a degree of the blur applied to the background portion 241.
[0105] After application of the blur and the blending, the modified third image data may be output (e.g., as fourth image data) by the multi-camera system as output image frames. For example, the processor 104 may execute a camera application 204 to instruct the display 114 to output the image frames.
[0106] The system 200 of Figure 2 may be configured to perform the operations described with reference to Figures 3 and 4 to determine output image frames 230. Figure 3 shows a flow chart of an example method 300 for processing image data for improved smooth alignment transition in a multi-camera system according to some embodiments of the disclosure. The processes shown in Figure 3 may obtain an improved smooth alignment transition in the outputted image frames based on image data captured by a multi-camera system (e.g., comprising a large baseline between cameras) , which results in outputted image frames with mitigated FOV jumps and reduced artifacts. Each of the operations described with reference to Figure 3 may be performed by one or a combination of the processor 104 (including cores 104A-N or AI engine 124) and / or the ISP 112.
[0107] At block 302, first image data and second image data is received from the first image sensor and the second image sensor, respectively. The first image data may have a larger FOV than the second image data. The first and second image data may be received, for example, from a bus coupled to the first camera 103 or from an analog front end (AFE) coupled to the first camera 103, and the second image data may be received, for example, from a bus coupled to the second camera 105 or from an analog front end (AFE) coupled to the second camera 105. For example, the first camera 103 and the second camera 105 may be part of a multi-camera system (e.g., a drone, an advanced driving assistance system (ADAS) , etc. ) . The first and second image data may alternatively be received from a wireless camera, in which the image data is received through one or more of the WAN adaptor 116a, the LAN adaptor 116b, and / or the PAN adaptor 116c. The first and second image data may alternatively be received from a memory location or a network storage location, such as when the image data was previously captured and is now retrieved from memory 106 and / or a remote location through one or more of the WAN adaptor 116a, the LAN adaptor 116b, and / or the PAN adaptor 116c. In some embodiments, the capture of image data may be initiated by a camera application executing on the processor 104, which causes camera control 210 to activate capture of image data by the first camera 103 and the second camera 105. The image data retrieved at block 302 may be then processed by the ISP 112 and / or processor 104 or other means for processing image data according to the operations described in one or more of the following blocks.
[0108] At block 304, a motion vector (e.g., motion vector 152) is determined. The motion vector may be based on the first image data and the second image data. In some aspects, the motion vector may be based on a zooming process (e.g., a request to zoom out of a previously output image data) . For example, a user of the multi-camera system may desire to zoom out relative to, a moving object captured by the relatively smaller FOV of an image frame captured by the second camera 105. A request may be received (e.g., by processor 104) to zoom out from an FOV of a previously output image data that is based on the FOV of the second image data. However, as the FOV of the second image data is relatively smaller than the FOV of the first image data, fulfilling the zoom out request may involve the first image data captured by the first camera 103. Method 300 may thus provide an improved SAT from the smaller FOV captured by the second camera 105 to the larger FOV captured by the first camera 103. The object, though also situated within the relatively larger FOV of the first image data, may not necessarily be at the center of the FOV of the first image data. The object, however, may be at the center of the FOV of the second image data. The motion vector may indicate a motion to be undertaken by the first camera 103 to align or adjust the FOV of the first camera 103 so that the object becomes situated at the center of the FOV of the first image data obtained by the first camera 103.
[0109] In some embodiments, the motion vector may be based on a sequence of image data received by one or both of the first image sensor (e.g., of the first camera 103) or the second image sensor (e.g., of the second camera 105) during the zooming process. For example, the motion vector may indicate a distance from the object on which the zooming is performed to the center of the FOV of the first image data. Alternatively, if the object is already situated at the center of the FOV of the second image data, the motion vector may indicate a distance between the center of the FOV of the second image data and the center of the FOV of the first image data.
[0110] As will be discussed in subsequent blocks, one or more output image frames may be determined based on the motion vector and a third image data generated using the second image data. In some embodiments, the one or more output image frames may be based on or predicated on the motion vector meeting one or more criteria. For example, the subsequent steps may be performed if the motion vector satisfies a predetermined threshold (e.g., if distance between the center of the FOV of the first image data and the object is above a predetermined threshold) . However, if the motion vector fails to satisfy the predetermined threshold, the multi-camera system may otherwise proceed differently with the zooming request (e.g., by performing conventional methods of zooming that may not achieve the improvements in smooth alignment transition described herein) .
[0111] At block 306, a center of the FOV of the first image data may be adjusted (e.g., recentered) based on the motion vector. For example, the center of the FOV of the first image data may be adjusted (e.g., via recentering functionality 208 of camera control (s) 210) by applying a non-central crop to the first image data. The non-central crop may indicate an area of the FOV (e.g., capturing the object) that is intended to become a new center for an updated FOV of the adjusted first image data. Furthermore, in some aspects, the non-central crop, though applied to the larger FOV of the first image data, may span an area covered by the FOV of the second image data. For example, the non-central crop may crop an area of the FOV of the first image data that approximates the smaller FOV of the second image data. Also or alternatively, a crop may be applied to the first image data, such that the crop is offset from the center of the FOV of the first image data by a distance that is based on (e.g., proportional to a magnitude of) the motion vector. In some embodiments, the center of the FOV of the first image data may be shifted towards a center of a crop applied to the first image data such that, prior to the shifting, the center of the crop may be offset from the center of the FOV of the first image data based on the motion vector.
[0112] In some embodiments, the adjustment of the center of the FOV of the first image data may further comprise scaling the first image data based on the zoom ratio of the second image data or scaling both the first image data and the second image data based on a desired zoom ratio (e.g., inputted by the user) . Also or alternatively, the scaling may be performed prior to applying a warp transformation, as described in block 308. In some embodiments, the adjusting and the scaling of the first image data based on the desired zoom ratio or based on the zoom ratio of the FOV of the second image data may be performed by applying the non-central crop to the first image data. For example, the non-central crop applied to the first image data to adjust the center of the FOV of the first image data at the object being zoomed into or out from may be based on the desired zoom ratio or zoom ratio of the second image data. Thus, the area of the non-central crop relative to the FOV of the first image data prior to the adjusting and scaling may be based on the zoom ratio.
[0113] At block 308, a warp transformation is applied to the second image data (e.g., via warping engine 138 of the ISP 112) . The warp transformation may be applied to render pixels of the FOV of the second image data to be more aligned or otherwise more similar to corresponding pixels of the first image data. For example, in some embodiments, the warp transformation may be applied by generating a warping matrix based on the adjusted FOV of the first image data and the FOV of the second image data. In some embodiments, the warp transformation of the second image data may result in third image data. The third image data may be determined by applying the warping matrix to the FOV of the second image data. For example, the warping matrix may map pixels of the second image data to corresponding pixels of the first image data (e.g., for comparison) . Objects or features within a pixel of the second image data may be adjusted to match, align, or otherwise become more similar to the objects or features within the corresponding pixel of the first image data. The adjustment may include but are not limited to a rotation of the pixel, object, or feature, a change in a perspective of the pixel, object, or feature, or a combination thereof. For example, objects or features may be stretched, compressed, or tilted. In some embodiments, the warp transformation may be based on a shooting angle of the image sensor, camera, or multi-camera system.
[0114] At block 310, the output image data may be determined based on the first image data and the second image data (e.g., processed and / or modified based on one or more of the aforementioned blocks of method 300) . In some embodiments, the output image data may be based on third image data formed from the second image data. For example, the third image data may be determined or generated based on the second image data after the warp transformation is applied to the second image data based on the adjusted and scaled first image data. Also or alternatively, the third image data may be generated based on the fusing of the first image data and the second image data. The outputted image frames of the output image data may achieve improvements in smooth alignment transition from the first and second image data obtained from the multi-camera system, by mitigating or minimizing jumps in FOV caused by camera switching during a zooming process. For example, by performing the adjusting and scaling of the first image data, or the warp transformations on the second image data, the aforementioned processes ensure that the output image frames of the output image data do not show disparities in image data captured from the first image sensor and the second image sensor during the zooming process.
[0115] As will be discussed in relation to Figure 4, further modifications can be performed on image data resulting from the aforementioned processing and modifications to the first and second image data (e.g., this image data) to determine output image data providing additional improvements to the smooth alignment transition in a multi-camera system. For example, as previously discussed varying levels of depth in scenes captured by multi-camera systems often negatively impact the user experience, as conventional multi-camera systems (particularly those with a larger baseline between cameras) cause FOV jumps in the outputted image frames. Various embodiments described in relation to Figure 4 address these shortcomings.
[0116] Figure 4 shows a flow chart of an example method 400 for processing image data of a multi-camera system for improved smooth alignment transition by using a bokeh effect to conceal or mitigate FOV jumps in the multi-camera system according to some embodiments of the disclosure. In some embodiments, method 400 include one or more example embodiments of block 310 of Figure 3, where the one or more output image frames are further determined based on a fourth image data generated using method 400. The processes shown in Figure 4 may obtain an improved smooth alignment transition in the outputted image frames based on image data captured by a multi-camera system (e.g., comprising a large baseline between cameras) , which results in outputted image frames with mitigated FOV jumps and reduced artifacts. Each of the operations described with reference to Figure 4 may be performed by one or a combination of the processor 104 (including cores 104A-N or AI engine 124) and / or the ISP 112.
[0117] At block 402, image data (third image data) may be determined or generated by applying the warp transformation to the second image data (e.g., as was applied in block 308 of method 300) . For example, as previously discussed, the warp transformation may be applied to render pixels of the FOV of the second image data to be more aligned or otherwise more similar to corresponding pixels of the first image data. In some embodiments, the third image data may comprise the second image data after the warp transformation is applied. Alternatively, the third image data may be determined or generated based on a fusion of the adjusted first image data and the warp transformed second image data.
[0118] At block 404, the third image data (e.g., obtained from block 402) may be segmented into portions. The portions may include at least a background portion and a foreground portion (e.g., background portion 241 and foreground portion 242) . In some aspects, the portion may also include a boundary portion (e.g., boundary portion 243) between the foreground portion and the background portion. Also or alternatively, the portions may further include any number of portions in addition to the foreground portion. The third image data may be segmented into the portions using depth information. For example, the segmented portions may be based on depth information 153 obtained via depth sensors 145. In some embodiments, an AI model (e.g., a machine learning model) may be used to determine the segments or portions by applying the third image data and / or depth information associated with the third image data into the AI model. For example, the segmentation engine 139 of the ISP 112 may rely on AI models 240 trained to determine depth information 153 based on image data, and / or trained to determine segments (e.g., foreground portion 242) based on the depth information 153 and / or the underlying image data. It is contemplated that objects or features in the third image data having less depth may be determined to be the foreground portion whereas the remaining portion may be determined to be a background portion. Also or alternatively, objects or features in the third image data having greater depth may be determined to be the background portion.
[0119] At block 406, a blur may be applied to the background portion of the third image data. For example, the ISP 112 may apply a blur (e.g., via the blurring engine 140) to pixels of the background portion 241. The applied blur may be a bokeh effect, a defocus blur, and / or Gaussian blur. In some embodiments, the blur may be applied to a pixel of the background portion by convolving the pixel with a kernel that represents a depth that is more than that of the foreground portion. In some embodiments, for example, where the segmented portions include a boundary portion between the background portion and the foreground portion, varying degrees of blur may also be applied to pixels of the background portion, as will be discussed in block 408.
[0120] At block 408, a fourth image data may be generated by blending a boundary between the foreground portion and the blurred background portion to generate a smoothened boundary in the fourth image data. For example, the ISP 112 may blend, via the blending engine 141, the boundary, such that output image frames based on the fourth image data reduce or erase any sharp boundary formed between the foreground and blurred background portions. In some aspects, the boundary may comprise a separately segmented boundary portion, which is subsequently blended. In some embodiments, the fourth image data may be generated by applying an alpha blending to the boundary portion of the third image data to generate the fourth image data having the smoothened boundary. For example, as a result of the alpha blending, pixels of the smoothened boundary farther from the foreground portion may have a greater degree of blur than pixels of the smoothened boundary closer to the foreground portion. However, pixels of the smoothened boundary farther from the foreground portion may nevertheless have a lesser degree of blur than a degree of the blur applied to the background portion. In some embodiments, applying the alpha blending to the boundary portion may include assigning weights to pixels of the boundary portion. Each weight may be associated with a proximity to the background portion and / or a distance from the foreground portion. Blur may be applied to each pixel of the boundary portion such that a degree of blur applied to a given pixel may be based on the respective weight assigned to the given pixel.
[0121] At block 410, the output image data may be determined based on the fourth image data. In some embodiments, block 410 may be an embodiment of block 310 of method 300 where the output image data is further based on the fourth image data generated using method 400. In some embodiments, the output image frames of the output image data may be frames of the fourth image data outputted by a multi-camera system, such as system 200. For example, the processor 104 may execute a camera application 204 to instruct the display 114 to output the image frames. Furthermore, the outputted image frames, which are generated (e.g., via methods 300 and 400) using image data from multiple cameras, may overcome issues with conventional smooth alignment transition, for example, by mitigating FOV jumps and reducing artifacts. Furthermore, the blending of the boundary between the foreground and the background portions ensures a smoother transition between the foreground and the background. Thus, the alignment and / or fusion of image data obtained from different image sensors (e.g., first image and the second image data) into third image data (e.g., using method 300) helps to mitigate any disparities in the image data captured from the first image sensor and the second image sensor during the zooming process.
[0122] Figure 5 is a block diagram illustrating an example process 500 for improved smooth alignment transition in a multi-camera system according to some embodiments of the disclosure. Process may be performed by one or more components of the multi-camera system, such as system 200. For example, the processor 104, ISP 112, or other processing circuitry, may be configured to perform one or more the methods or steps of Figure 5.
[0123] Process 500 may begin with receiving a zoom request (block 502) . For example, the zoom request may be received by the processor 104 (e.g., based on user input) , and may be a request to zoom out from a current zoom ratio in an image data stream to a desired zoom ratio. The image data stream may be formed by image data frames output by a multi-camera system. The image data stream may thus be modified and adapted during the zooming process based on the methods described herein, to provide improved smooth alignment transition. The multi-camera system may include at least the first image sensor 101 (e.g., of the first camera 103) capturing image data (second image data) having a relatively larger FOV and the second image sensor 102 (e.g., of the second camera 105) capturing image data (second image data) having a relatively smaller FOV. The zoom out request may involve a transition from the relatively smaller FOV (e.g., of the second camera 105) to the relatively larger FOV (e.g., of the first camera 103) .
[0124] Furthermore, the zoom request may be associated with a specific location or object in the FOV of the output image data stream. For example, a request to zoom out may involve zooming out relative to an object shown within a relatively smaller FOV in the output image data stream in order to view the object from a relatively larger FOV. However, the object that is the subject of the zooming request may not necessarily be at the center of the relatively larger FOV of the first image data captured by the first image sensor. In some embodiments, the zoom request may be based on a user input causing movement of the second camera 105 so that the object on which the zooming procedure is desired to be performed becomes situated within the center of the FOV of the second image data obtained from the second image sensor of the second camera 105. For example, the object may be away from the center of the larger FOV of the first image data even though the object may be at the center of the FOV of the second image data captured by the second image sensor, after the adjusting the second camera to move towards capturing the object at the center of the FOV of the second image data.
[0125] At block 510, a motion vector may be determined based on the zoom request. For example, the motion vector 152 may be based on whether the larger FOV of the first image data (e.g., obtained by the first camera 103) is centered at the object in which the zoom-in or zoom-out is being performed. If the FOV of the first image data is not centered at the object, the motion vector 152 may represent the distance between the center of the FOV of the first image data and the object on which the zooming out from is to be performed. Also or alternatively, as the smaller FOV of the second image data (e.g., obtained by the second camera 105) may be centered at the object on which the zooming in to or zooming out from is being requested, the motion vector 152 may represent the distance from the center of the larger FOV of the first image data to the center of the FOV of the second image data.
[0126] At block 512, the multi-camera system (e.g., processor 104 of system 200) may determine whether the motion vector satisfies a predetermined threshold. For example, the multi-camera system may determine whether the magnitude of the motion vector (e.g., the distance between the center of the FOV of the first image data and the object) is above a predetermined threshold. Subsequent steps of process 500 (e.g., blocks 516 through 540) for improving smooth alignment transition may be performed if the motion vector satisfies the predetermined threshold (e.g., the magnitude is above the predetermined threshold) . However, if the motion vector fails to satisfy the predetermined threshold, the multi-camera system may otherwise proceed differently with the zooming request (e.g., by performing conventional methods of zooming that may not achieve the improvements in smooth alignment transition described herein) , such as by proceeding to execute the zoom request (block 514) without the processing at block 516 and subsequent blocks.
[0127] At block 516, if determined that the motion vector satisfies the predetermined threshold, the multi-camera system may adjust the center of (e.g., recenter) the FOV of the first image data based on the motion vector. For example, the FOV of the first image data may be adjusted so that the object on which the zoom request is performed becomes situated at the center of the adjusted FOV of the first image data. At block 518, the multi-camera system may scale the first image data to a zoom ratio of the second image data. For example, the second camera capturing the second image data may be used to select an area encompassing (e.g., as the FOV of the second image data) the object on which the zoom request is to be performed. By scaling the adjusted (e.g., recentered) first image data to the zoom ratio of the second image data, the adjusted first image data may be scaled to a zoom ratio desired by the zoom request.
[0128] In some embodiments, the adjusting and scaling of the first image data based on the motion vector and the zoom ratio of the second image data may be performed by applying a non-center crop to the FOV of the first image data. The non-central crop, though applied to the relatively larger FOV of the first image data, may span an area covered by the relatively smaller FOV of the second image data. For example, the non-central crop may crop an area of the FOV of the first image data that approximates the smaller FOV of the second image data. Thus, by applying the non-central crop to the first image data to encompass the object in the first image data that is the subject of the zoom request, the first image data is cropped to form an updated FOV of the first image data having the object at the center and having the desired zoom ratio reflected by the updated FOV. Also or alternatively, a crop may be applied to the first image data, such that the crop is adjusted from the center of the FOV of the first image data by a distance that is based on (e.g., proportional to a magnitude of) the motion vector. In some embodiments, the center of the FOV of the first image data may be shifted towards a center of a crop applied to the first image data such that, prior to the shifting, the center of the crop may be offset from the center of the FOV of the first image data based on the motion vector.
[0129] At block 520, the multi-camera system may warp transform the second image data to adjust the FOV of the second image data towards the adjusted and scaled FOV of the first image data. For example, pixels of the FOV of the second image data may be adjusted (e.g., via rotation, perspective changes, or a combination thereof) to align, match, or otherwise become more similar to corresponding pixels of the adjusted and scaled FOV of the first image data. In some embodiments, the warp transformation may be performed via a warping matrix 522 configured to map pixels of the FOV of the second image data with corresponding pixels of the FOV of the first image data. The result of the processes of the adjusting and scaling of the first image data and the warp transformation of the FOV of the second image data based on the adjusted and scaled FOV of the first image data may include an alignment 524 of the first image data with the second image data. In some embodiments, the alignment 524 may be manifested as third image data (also referred to as “aligned image data” ) generated based on the modified second image data (e.g., after the warp transformation) . Also or alternatively, the alignment 524 may be manifested as third image data generated based on a fusion of the modified second image data with the adjusted and scaled first image data.
[0130] At block 526, the multi-camera system may segment the third image data based on depth to differentiate between foreground and background portions of the third image data. For example, the segmented portions may be based on depth information 153 obtained via depth sensors 145. In some embodiments, an AI model (e.g., a machine learning model) may be used to determine the segments or portions by applying the third image data and / or depth information associated with the third image data into the AI model. For example, the segmentation engine 139 of the ISP 112 may rely on AI models 240 trained to determine depth information 153 based on image data, and / or trained to determine segments (e.g., foreground portion 242) based on the depth information 153 and / or the underlying image data. It is contemplated that objects or features in the third image data having less depth may be determined to be the foreground portion whereas the remaining portion may be determined to be a background portion. Also or alternatively, objects or features in the third image data having greater depth may be determined to be the background portion.
[0131] At block 528, the multi-camera system may apply a blur to the background portion. For example, the ISP 112 may apply a blur (e.g., via the blurring engine 140) to pixels of the background portion 243. The applied blur may be a bokeh effect, a defocus blur, and / or Gaussian blur. In some embodiments, the blur may be applied to a pixel of the background portion by convolving the pixel with a kernel that represents a depth that is more than that of the foreground portion.
[0132] At block 530, the boundary between the foreground portion and the blurred background portion may be refined via blending. The blending may generate a smoothened boundary in the outputted image frames 550. For example, the ISP 112 may blend, via the blending engine 141, the boundary, such that output image frames exhibit reduced or eliminated boundaries between the foreground and blurred background portions. In some aspects, the boundary may comprise a separately segmented boundary portion, which is subsequently blended at block 530. In some embodiments, the blending may be based on an alpha blending technique, where a property (e.g., a degree of blur) of pixels in the background portion may be combined or averaged with a property (e.g., a degree of blur) of pixels in the foreground portion to a property (e.g., a degree of blur) for pixels in the boundary portion to generate the smoothened boundary. In some embodiments, the blending may cause pixels of the smoothened boundary farther from the foreground portion to have a greater degree of blur than pixels of the smoothened boundary closer to the foreground portion. However, pixels of the smoothened boundary farther from the foreground portion may nevertheless have less blur than a degree of the blur applied to the background portion.
[0133] At block 540, image frames may be output based on the based on the image data modified under the blocks. The modified image data from which the one or more output image frames 550 are output may be referred to as fourth image data. For example, the processor 104 may execute a camera application 204 to instruct the display 114 to output the image frames 550. The disclosed techniques for generating the outputted image frames overcoming issues with conventional smooth alignment transition, for example, by mitigating FOV jumps and reducing artifacts. For example, the blending of the boundary between the foreground and the background portions ensures a smoother transition between the foreground and the background. Furthermore, the alignment and / or fusion of image data obtained from different image sensors (e.g., first image and the second image data) into third image data help to mitigate any disparities in the image data captured from the first image sensor 101 and the second image sensor 102 during the zooming process. The output image frames at block 540 may be used as part of a preview operation in which the device presents the scene captured by the image while waiting for the user to press a capture button.
[0134] In one or more aspects, techniques for supporting image processing may include additional aspects, such as any single aspect or any combination of aspects described below or in connection with one or more other processes or devices described elsewhere herein. In a first aspect, supporting image processing may include an apparatus comprising a memory; and one or more processors coupled to the memory. The one or more processors are configured to: receive, from a first image sensor, first image data; receive, from a second image sensor, second image data, wherein the first image data has a larger field of view (FOV) than the FOV of the second image data; determine, based on the first image data and the second image data, a motion vector; adjust, based on the motion vector, the FOV of the first image data; apply a warp transformation to the second image data, wherein the warp transformation adjusts the FOV of the second image data towards the offset FOV of the first image data; and determine output image data based on the first image data and the second image data.
[0135] Additionally, the apparatus may perform or operate according to one or more aspects as described below. In some implementations, the apparatus includes a remote server, such as a cloud-based computing solution, which receives image data for processing to determine output image frames. In some implementations, the apparatus may include at least one processor, and a memory coupled to the processor. The processor may be configured to perform operations described herein with respect to the apparatus. In some other implementations, the apparatus may include a non-transitory computer-readable medium having program code recorded thereon and the program code may be executable by a computer for causing the computer to perform operations described herein with reference to the apparatus. In some implementations, the apparatus may include one or more means configured to perform operations described herein. In some implementations, a method of image processing may include one or more operations described herein with reference to the apparatus.
[0136] In a second aspect, in combination with the first aspect, the one or more processors are configured to adjust the FOV of the first image data, apply the warp transformation, and determine the output image data based on the motion vector meeting one or more criteria.
[0137] In a third aspect, in combination with one or more of the first aspect or the second aspect, the one or more processors are configured to adjust the FOV of the first image data by: shifting a center of the FOV of the first image data towards a center of a crop applied to the first image data, wherein, prior to the shifting, the center of the crop is offset from the center of the FOV of the first image data based on the motion vector.
[0138] In a fourth aspect, in combination with one or more of the first aspect through the third aspect, the one or more processors are further configured to: prior to applying the warp transformation, scaling the first image data based on a zoom ratio of the FOV of the second image data.
[0139] In a fifth aspect, in combination with one or more of the first aspect through the fourth aspect, the one or more processors are further configured to: prior to determining the motion vector, receive a request to zoom out from an FOV of a previously output image data based on the FOV of the second image data; and determine the motion vector in response to the request, wherein the motion vector indicates a distance between a center of the FOV of the second image data and the center of the FOV of the first image data.
[0140] In a sixth aspect, in combination with one or more of the first aspect through the fifth aspect, the one or more processors are further configured to apply the warp transformation by:generating a warping matrix based on the offset FOV of the first image data and the FOV of the second image data; and determining a third image data by applying the warping matrix to the FOV of the second image data, wherein the output image data is further based on the third image data.
[0141] In a seventh aspect, in combination with one or more of the first aspect through the sixth aspect, the one or more processors are further configured to determine the output image data by: segmenting the third image data into a plurality of portions comprising a foreground portion and a background portion using depth information associated with the third image data; applying a blur to the background portion of the third image data to generate a blurred background portion; and generating, based on the third image data, a fourth image data by blending a boundary between the foreground portion and the blurred background portion, wherein the output image data is further based on the fourth image data.
[0142] In an eighth aspect, in combination with one or more of the first aspect through the seventh aspect, the one or more processors are further configured to segment the third image data by:determining, by applying the third image data into a machine learning model, the depth information associated with the third image data.
[0143] In a ninth aspect, in combination with one or more of the first aspect through the eighth aspect, the plurality of portions further includes a boundary portion between the background portion and the foreground portion, wherein the one or more processors are further configured to generate the fourth image data by: applying an alpha blending to the boundary portion of the third image data, wherein, after the application of the alpha blending, pixels of the boundary portion farther from the foreground portion have a greater degree of blur than pixels of the boundary portion closer to the foreground portion, but have a lesser degree of blur than a degree of the blur applied to the background portion.
[0144] In a tenth aspect, an image capture device is disclosed that may perform or operate according to one or more aspects as described below. In some implementations, the image capture device includes: a first image sensor configured to capture a first field of view (FOV) ; a second image sensor configured to capture a second FOV, wherein the first FOV is larger than the second FOV; a memory; and one or more processors coupled to the memory. In some implementations, the one or more processor are configured to: receive, from the first image sensor, first image data; receive, from the second image sensor, second image data; determine, based on the first image data and the second image data, a motion vector; adjust, based on the motion vector, the FOV of the first image data; apply a warp transformation to the second image data to generate a third image data, wherein the warp transformation adjusts the FOV of the second image data towards the adjusted FOV of the first image data; segment the third image data into a plurality of portions comprising a foreground portion and a background portion using depth information associated with the third image data; apply a blur to the background portion of the third image data to generate a blurred background portion; generate, based on the third image data, a fourth image data by blending a boundary between the foreground portion and the blurred background portion; and determining output image data based on the fourth image data.
[0145] In an eleventh aspect, in combination with the tenth aspect, the plurality of portions further includes a boundary portion between the background portion and the foreground portion. In some implementations, the one or more processors are further configured to generate the fourth image data by: applying an alpha blending to the boundary portion of the third image data to generate the fourth image data, wherein, after application of the alpha blending, pixels of the boundary portion farther from the foreground portion have a greater degree of blur than pixels of the boundary portion closer to the foreground portion, but have a lesser degree of blur than a degree of the blur applied to the background portion.
[0146] In a twelfth aspect, a method is disclosed according to one or more aspects as described below. In some implementations, the method includes: receiving, from a first image sensor, first image data; receiving, from a second image sensor, second image data, wherein the first image data has a larger field of view (FOV) than the FOV of the second image data; determining, based on the first image data and the second image data, a motion vector; adjusting, based on the motion vector, the FOV of the first image data; applying a warp transformation to the second image data, wherein the warp transformation adjusts the FOV of the second image data towards the adjusted FOV of the first image data; and determining output image data based on the first image data and the second image data.
[0147] In a thirteenth aspect, in combination with the twelfth aspect, adjusting the FOV of the first image data, applying the warp transformation, and determining the output image data is based on the motion vector meeting one or more criteria.
[0148] In a fourteenth aspect, in combination with twelfth aspect or the thirteenth aspect, adjusting the FOV of the first image data includes: shifting a center of the FOV of the first image data towards a center of a crop applied to the first image data, wherein, prior to the shifting, the center of the crop is offset from the center of the FOV of the first image data based on the motion vector.
[0149] In a fifteenth aspect, in combination with one or more of the twelfth aspect through the fourteenth aspect, the method further includes: prior to applying the warp transformation, scaling the first image data based on a zoom ratio of the FOV of the second image data.
[0150] In a sixteenth aspect, in combination with one or more of the twelfth aspect through the fifteenth aspect, the method further includes: prior to determining the motion vector, receiving a request to zoom out from an FOV of a previously output image data based on the FOV of the second image data; and determining the motion vector in response to the request, wherein the motion vector indicates a distance between a center of the FOV of the second image data and the center of the FOV of the first image data.
[0151] In a seventeenth aspect, in combination with one or more of the twelfth aspect through the sixteenth aspect, applying the warp transformation includes: generating a warping matrix based on the adjusted FOV of the first image data and the FOV of the second image data; and determining a third image data by applying the warping matrix to the FOV of the second image data, wherein the output image data is further based on the third image data.
[0152] In an eighteenth aspect, in combination with one or more of the twelfth aspect through the seventeenth aspect, determining the output image data further includes: segmenting the third image data into a plurality of portions comprising a foreground portion and a background portion using depth information associated with the third image data; applying a blur to the background portion of the third image data to generate a blurred background portion; and generating, based on the third image data, a fourth image data by blending a boundary between the foreground portion and the blurred background portion, wherein the output image data is further based on the fourth image data.
[0153] In a nineteenth aspect, in combination with one or more of the first aspect through the eighteenth aspect, segmenting the third image data includes: determining, by applying the third image data into a machine learning model, the depth information associated with the third image data.
[0154] In a twentieth aspect, in combination with one or more of the first aspect through the nineteenth aspect, the plurality of portions further includes a boundary portion between the background portion and the foreground portion, wherein generating the fourth image data includes: applying an alpha blending to the boundary portion of the third image data, wherein, after the application of the alpha blending, pixels of the boundary portion farther from the foreground portion have a greater degree of blur than pixels of the boundary portion closer to the foreground portion, but have a lesser degree of blur than a degree of the blur applied to the background portion.
[0155] In the figures, a single block may be described as performing a function or functions. The function or functions performed by that block may be performed in a single component or across multiple components, and / or may be performed using hardware, software, or a combination of hardware and software. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps are described below generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure. Also, the example devices may include components other than those shown, including well-known components such as a processor, memory, and the like.
[0156] Aspects of the present disclosure are applicable to any electronic device including, coupled to, or otherwise processing data from one, two, or more image sensors capable of capturing image frames (or “frames” ) . The terms “output image frame, ” “modified image frame, ” and “corrected image frame” may refer to an image frame that has been processed by any of the disclosed techniques to adjust raw image data received from an image sensor. Further, aspects of the disclosed techniques may be implemented for processing image data received from image sensors of the same or different capabilities and characteristics (such as resolution, shutter speed, or sensor type) . Further, aspects of the disclosed techniques may be implemented in devices for processing image data, whether or not the device includes or is coupled to image sensors. For example, the disclosed techniques may include operations performed by processing devices in a cloud computing system that retrieve image data for processing that was previously recorded by a separate device having image sensors.
[0157] Unless specifically stated otherwise as apparent from the following discussions, it is appreciated that throughout the present application, discussions using terms such as “accessing, ” “receiving, ” “sending, ” “using, ” “selecting, ” “determining, ” “normalizing, ” “multiplying, ” “averaging, ” “monitoring, ” “comparing, ” “applying, ” “updating, ” “measuring, ” “deriving, ” “settling, ” “generating, ” or the like, refer to the actions and processes of a computer system, or similar electronic computing device, that manipulates and transforms data represented as physical (electronic) quantities within the computer system’s registers and memories into other data similarly represented as physical quantities within the computer system’s registers, memories, or other such information storage, transmission, or display devices. The use of different terms referring to actions or processes of a computer system does not necessarily indicate different operations. For example, “determining” data may refer to “generating” data. As another example, “determining” data may refer to “retrieving” data.
[0158] The terms “device” and “apparatus” are not limited to one or a specific number of physical objects (such as one smartphone, one camera controller, one processing system, and so on) . As used herein, a device may be any electronic device with one or more parts that may implement at least some portions of the disclosure. While the description and examples herein use the term “device” to describe various aspects of the disclosure, the term “device” is not limited to a specific configuration, type, or number of objects. As used herein, an apparatus may include a device or a portion of the device for performing the described operations.
[0159] Certain components in a device or apparatus described as “means for accessing, ” “means for receiving, ” “means for sending, ” “means for using, ” “means for selecting, ” “means for determining, ” “means for normalizing, ” “means for multiplying, ” or other similarly-named terms referring to one or more operations on data, such as image data, may refer to processing circuitry (e.g., application specific integrated circuits (ASICs) , digital signal processors (DSP) , graphics processing unit (GPU) , central processing unit (CPU) , computer vision processor (CVP) , or neural signal processor (NSP) ) configured to perform the recited function through hardware, software, or a combination of hardware configured by software.
[0160] Those of skill in the art would understand that information and signals may be represented using any of a variety of different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips that may be referenced throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof.
[0161] Components, the functional blocks, and the modules described herein with respect to the Figures referenced above include processors, electronics devices, hardware devices, electronics components, logical circuits, memories, software codes, firmware codes, among other examples, or any combination thereof. Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, application, software applications, software packages, routines, subroutines, objects, executables, threads of execution, procedures, and / or functions, among other examples, whether referred to as software, firmware, middleware, microcode, hardware description language or otherwise. In addition, features discussed herein may be implemented via specialized processor circuitry, via executable instructions, or combinations thereof.
[0162] Those of skill in the art will understand that one or more blocks (or operations) described with reference to Figures 3, 4, and 5 may be combined with one or more blocks (or operations) described with reference to another of the figures. For example, one or more blocks (or operations) of Figure 3 and 5 may be combined with one or more blocks (or operations) of Figures 1-2. As another example, one or more blocks associated with Figure 4 and 5 may be combined with one or more blocks (or operations) associated with Figures 1-2.
[0163] Those of skill in the art would further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the disclosure herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure. Skilled artisans will also readily recognize that the order or combination of components, methods, or interactions that are described herein are merely examples and that the components, methods, or interactions of the various aspects of the present disclosure may be combined or performed in ways other than those illustrated and described herein.
[0164] The various illustrative logics, logical blocks, modules, circuits and algorithm processes described in connection with the implementations disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. The interchangeability of hardware and software has been described generally, in terms of functionality, and illustrated in the various illustrative components, blocks, modules, circuits, and processes described above. Whether such functionality is implemented in hardware or software depends upon the particular application and design constraints imposed on the overall system.
[0165] The hardware and data processing apparatus used to implement the various illustrative logics, logical blocks, modules and circuits described in connection with the aspects disclosed herein may be implemented or performed with a general purpose single-or multi-chip processor, a digital signal processor (DSP) , an application specific integrated circuit (ASIC) , a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general-purpose processor may be a microprocessor, or, any conventional processor, controller, microcontroller, or state machine. In some implementations, a processor may be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. In some implementations, particular processes and methods may be performed by circuitry that is specific to a given function.
[0166] In one or more aspects, the functions described may be implemented in hardware, digital electronic circuitry, computer software, firmware, including the structures disclosed in this specification and their structural equivalents thereof, or in any combination thereof. Implementations of the subject matter described in this specification also may be implemented as one or more computer programs, which is one or more modules of computer program instructions, encoded on a computer storage media for execution by, or to control the operation of, data processing apparatus.
[0167] If implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. The processes of a method or algorithm disclosed herein may be implemented in a processor-executable software module which may reside on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that may be enabled to transfer a computer program from one place to another. A storage media may be any available media that may be accessed by a computer. By way of example, and not limitation, such computer-readable media may include random-access memory (RAM) , read-only memory (ROM) , electrically erasable programmable read-only memory (EEPROM) , CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that may be used to store desired program code in the form of instructions or data structures and that may be accessed by a computer. Also, any connection may be properly termed a computer-readable medium. Disk and disc, as used herein, includes compact disc (CD) , laser disc, optical disc, digital versatile disc (DVD) , floppy disk, and Blu-ray disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media. Additionally, the operations of a method or algorithm may reside as one or any combination or set of codes and instructions on a machine readable medium and computer-readable medium, which may be incorporated into a computer program product.
[0168] Various modifications to the implementations described in this disclosure may be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to some other implementations without departing from the spirit or scope of this disclosure. Thus, the claims are not intended to be limited to the implementations shown herein but are to be accorded the widest scope consistent with this disclosure, the principles and the novel features disclosed herein.
[0169] Additionally, a person having ordinary skill in the art will readily appreciate, opposing terms such as “upper” and “lower, ” or “front” and back, ” or “top” and “bottom, ” or “forward” and “backward” are sometimes used for ease of describing the figures, and indicate relative positions corresponding to the orientation of the figure on a properly oriented page, and may not reflect the proper orientation of any device as implemented.
[0170] Certain features that are described in this specification in the context of separate implementations also may be implemented in combination in a single implementation. Conversely, various features that are described in the context of a single implementation also may be implemented in multiple implementations separately or in any suitable subcombination. Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination may in some cases be excised from the combination, and the claimed combination may be directed to a subcombination or variation of a subcombination.
[0171] Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown, or in sequential order, or that all illustrated operations be performed to achieve desirable results. Further, the drawings may schematically depict one or more example processes in the form of a flow diagram. However, other operations that are not depicted may be incorporated in the example processes that are schematically illustrated. For example, one or more additional operations may be performed before, after, simultaneously, or between any of the illustrated operations. In certain circumstances, multitasking and parallel processing may be advantageous. Moreover, the separation of various system components in the implementations described above should not be understood as requiring such separation in all implementations, and it should be understood that the described program components and systems may generally be integrated together in a single software product or packaged into multiple software products. Additionally, some other implementations are within the scope of the following claims. In some cases, the actions recited in the claims may be performed in a different order and still achieve desirable results.
[0172] As used herein, including in the claims, the term “or, ” when used in a list of two or more items, means that any one of the listed items may be employed by itself, or any combination of two or more of the listed items may be employed. For example, if a composition is described as containing components A, B, or C, the composition may contain A alone; B alone; C alone; A and B in combination; A and C in combination; B and C in combination; or A, B, and C in combination. Also, as used herein, including in the claims, “or” as used in a list of items prefaced by “at least one of” indicates a disjunctive list such that, for example, a list of “at least one of A, B, or C” means A or B or C or AB or AC or BC or ABC (that is A and B and C) or any of these in any combination thereof.
[0173] The term “substantially” is defined as largely, but not necessarily wholly, what is specified (and includes what is specified; for example, substantially 90 degrees includes 90 degrees and substantially parallel includes parallel) , as understood by a person of ordinary skill in the art. In any disclosed implementations, the term “substantially” may be substituted with “within [apercentage] of” what is specified, where the percentage includes . 1, 1, 5, or 10 percent.
[0174] The previous description of the disclosure is provided to enable any person skilled in the art to make or use the disclosure. Various modifications to the disclosure will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other variations without departing from the spirit or scope of the disclosure. Thus, the disclosure is not intended to be limited to the examples and designs described herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1.A method, comprising:receiving, from a first image sensor, first image data;receiving, from a second image sensor, second image data, wherein the first image data has a field of view (FOV) larger than a FOV of the second image data;determining, based on the first image data and the second image data, a motion vector;adjusting, based on the motion vector, the FOV of the first image data such that the first image data has an adjusted FOV;applying a warp transformation to the second image data, to adjust the FOV of the second image data towards the adjusted FOV of the first image data; anddetermining output image data based on the first image data and the second image data after adjusting the FOV of the first image data and applying the warp transformation.2.The method of claim 1, wherein adjusting the FOV of the first image data, applying the warp transformation, and determining the output image data is based on the motion vector meeting one or more criteria.3.The method of claim 1, wherein adjusting the FOV of the first image data comprises:shifting a center of the FOV of the first image data towards a center of a crop applied to the first image data, wherein, prior to the shifting, the center of the crop is offset from the center of the FOV of the first image data based on the motion vector.4.The method of claim 1, further comprising:prior to applying the warp transformation, scaling the first image data based on a zoom ratio of the FOV of the second image data.5.The method of claim 1, further comprising:prior to determining the motion vector, receiving a request to zoom out from an FOV of a previously output image data based on the FOV of the second image data; anddetermining the motion vector in response to the request, wherein the motion vector indicates a distance between a center of the FOV of the second image data and the center of the FOV of the first image data.6.The method of claim 1, wherein applying the warp transformation comprises:generating a warping matrix based on the adjusted FOV of the first image data and the FOV of the second image data; anddetermining a third image data by applying the warping matrix to the FOV of the second image data, wherein the output image data is further based on the third image data.7.The method of claim 6, wherein determining the output image data further comprises:segmenting the third image data into a plurality of portions comprising a foreground portion and a background portion using depth information associated with the third image data;applying a blur to the background portion of the third image data to generate a blurred background portion; andgenerating, based on the third image data, a fourth image data by blending a boundary between the foreground portion and the blurred background portion, wherein the output image data is further based on the fourth image data.8.The method of claim 7, wherein segmenting the third image data comprises:determining, by applying the third image data into a machine learning model, the depth information associated with the third image data.9.The method of claim 7, wherein the plurality of portions further comprises a boundary portion between the background portion and the foreground portion, wherein generating the fourth image data comprises:applying an alpha blending to the boundary portion of the third image data such that pixels of the boundary portion farther from the foreground portion have a greater degree of blur than pixels of the boundary portion closer to the foreground portion and have a lesser degree of blur than a degree of the blur applied to the background portion.10.An apparatus, comprising:a memory; andone or more processors coupled to the memory, the one or more processors being configured to:receive, from a first image sensor, first image data;receive, from a second image sensor, second image data, wherein the first image data has a larger field of view (FOV) than the FOV of the second image data;determine, based on the first image data and the second image data, a motion vector;adjusting, based on the motion vector, the FOV of the first image data such that the first image data has an adjusted FOV;apply a warp transformation to the second image data, wherein the warp transformation adjusts the FOV of the second image data towards the larger FOV of the first image data; anddetermine output image data based on the first image data and the second image data.11.The apparatus of claim 10, wherein the one or more processors are configured to adjust the FOV of the first image data, apply the warp transformation, and determine the output image data based on the motion vector meeting one or more criteria.12.The apparatus of claim 10, wherein the one or more processors are configured to adjust the FOV of the first image data by:shifting a center of the FOV of the first image data towards a center of a crop applied to the first image data, wherein, prior to the shifting, the center of the crop is offset from the center of the FOV of the first image data based on the motion vector.13.The apparatus of claim 10, wherein the one or more processors are further configured to:prior to applying the warp transformation, scale the first image data based on a zoom ratio of the FOV of the second image data.14.The apparatus of claim 10, wherein the one or more processors are further configured to:prior to determining the motion vector, receive a request to zoom out from an FOV of a previously output image data based on the FOV of the second image data; anddetermine the motion vector in response to the request, wherein the motion vector indicates a distance between a center of the FOV of the second image data and the center of the FOV of the first image data.15.The apparatus of claim 10, wherein the one or more processors are further configured to apply the warp transformation by:generating a warping matrix based on the adjusted FOV of the first image data and the FOV of the second image data; anddetermining a third image data by applying the warping matrix to the FOV of the second image data, wherein the output image data is further based on the third image data.16.The apparatus of claim 15, wherein the one or more processors are further configured to determine the output image data by:segmenting the third image data into a plurality of portions comprising a foreground portion and a background portion using depth information associated with the third image data;applying a blur to the background portion of the third image data to generate a blurred background portion; andgenerating, based on the third image data, a fourth image data by blending a boundary between the foreground portion and the blurred background portion, wherein the output image data is further based on the fourth image data.17.The apparatus of claim 16, wherein the one or more processors are further configured to segment the third image data by:determining, by applying the third image data into a machine learning model, the depth information associated with the third image data.18.The apparatus of claim 16, wherein the plurality of portions further comprises a boundary portion between the background portion and the foreground portion, wherein the one or more processors are further configured to generate the fourth image data by:applying an alpha blending to the boundary portion of the third image data such that pixels of the boundary portion farther from the foreground portion have a greater degree of blur than pixels of the boundary portion closer to the foreground portion and have a lesser degree of blur than a degree of the blur applied to the background portion.19.An image capture device, comprising:a first image sensor configured to capture a first field of view (FOV) ;a second image sensor configured to capture a second FOV, wherein the first FOV is larger than the second FOV;a memory; andone or more processors coupled to the memory, the one or more processor being configured to:receive, from the first image sensor, first image data;receive, from the second image sensor, second image data;determine, based on the first image data and the second image data, a motion vector;adjust, based on the motion vector, the FOV of the first image data;apply a warp transformation to the second image data to generate a third image data, wherein the warp transformation adjusts the FOV of the second image data towards the adjusted FOV of the first image data;segment the third image data into a plurality of portions comprising a foreground portion and a background portion using depth information associated with the third image data;apply a blur to the background portion of the third image data to generate a blurred background portion;generate, based on the third image data, a fourth image data by blending a boundary between the foreground portion and the blurred background portion; anddetermining output image data based on the fourth image data.20.The image capture device of claim 19, wherein the plurality of portions further comprises a boundary portion between the background portion and the foreground portion, wherein the one or more processors are further configured to generate the fourth image data by:applying an alpha blending to the boundary portion of the third image data to generate the fourth image data, wherein, after application of the alpha blending, pixels of the boundary portion farther from the foreground portion have a greater degree of blur than pixels of the boundary portion closer to the foreground portion, but have a lesser degree of blur than a degree of the blur applied to the background portion.