Method and system for searching images using rotation gesture input

The system allows users to change the perspective of 2D images using rotational gestures and crowdsourcing, addressing the lack of angle-specific viewing in image search engines by dynamically updating the image database with user-contributed content.

JP7578366B2Active Publication Date: 2024-11-06INTERNATIONAL BUSINESS MACHINE CORPORATION
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
JP2022527719
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2019-11-20
Filing Date
2020-11-16
Publication Date
2024-11-06
Estimated Expiration
2040-11-16

AI Technical Summary

Technical Problem

Image search engines do not provide 2D image results based on a specific viewing angle of objects, and users cannot specify a desired angle for viewing 2D images without converting them to 3D models.

Method used

A system that allows users to input a desired change in perspective of a 2D image using rotational gestures, searches for and displays 2D images matching the new perspective, and utilizes crowdsourcing to gather missing viewpoints.

Benefits of technology

Enables users to view 2D images from different angles without converting to 3D models, providing a seamless experience and continuously improving the image database with user-contributed content.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007578366000001
    Figure 0007578366000001
  • Figure 0007578366000002
    Figure 0007578366000002
  • Figure 0007578366000003
    Figure 0007578366000003
Patent Text Reader

Abstract

A method for searching images using rotational gesture input includes identifying an object in a two-dimensional reference image, determining a three-dimensional reference orientation axis of the object based on at least one attribute of the object, receiving an input requesting a change in the three-dimensional perspective of the object, determining a new orientation axis based on the input and the reference orientation axis, performing a search query based on the new orientation axis and the at least one attribute of the object on a set of two-dimensional images, and displaying image search results ranked based on correlation with the new orientation axis and the attribute of the object.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] FIELD OF THE DISCLOSURE This disclosure relates to image searching, and more particularly, to searching for images using rotation gesture input. [Background technology]

[0002] A display device can display an object in two dimensions ("2D") or three dimensions ("3D"). When an object is displayed in 2D (e.g., as a flat photo of a car), a computing system lacks the information necessary to change the view of the object to a different angular perspective. However, an object displayed in 3D can be viewed on a two-dimensional display device in a variety of different angular orientations or perspectives. For example, a computer-aided design ("CAD") program allows a user to rotate a solid (or wireframe) representation of an object to view the object from a different angular perspective. The three-dimensional model of the object contains additional information (e.g., X, Y, and Z axis information) that allows the program to regenerate a new image of the object each time the user requests a change of perspective. However, for an object viewed in 2D, a rotation gesture (or other form of user input) does not result in a different viewing angle of the object. Therefore, the user cannot provide input to change the viewed perspective of a rotation of the 2D image. Therefore, there is a need to address the problems noted above in the prior art. Summary of the Invention

[0003] Viewed from a first aspect, the present invention provides a method for searching images, the method comprising: identifying a reference object in a two-dimensional reference image; determining a three-dimensional reference orientation axis of the reference object based on at least one attribute of the reference object; receiving an input requesting a change in the three-dimensional perspective of the reference object; determining a new orientation axis based on the input and the reference orientation axis; performing a search query on a set of two-dimensional images based on the new orientation axis and the at least one attribute of the reference object; and displaying image search results ranked based on correlation with the new orientation axis and the at least one attribute of the reference object.

[0004] Viewed from a further aspect, the present invention provides a computer system comprising a computer-readable storage medium having program instructions stored thereon and one or more processors configured to execute the program instructions to perform a method, the method including: identifying a reference object in a two-dimensional reference image; determining a three-dimensional reference orientation axis of the reference object based on at least one attribute of the reference object; receiving an input requesting a change in the three-dimensional perspective of the reference object; determining a new orientation axis based on the input and the reference orientation axis; performing a search query on the set of two-dimensional images based on the new orientation axis and the at least one attribute of the reference object; and displaying image search results ranked based on correlation with the new orientation axis and the at least one attribute of the reference object.

[0005] Viewed from a further aspect, the present invention provides a computer program product for searching for images, the computer program product comprising a computer readable storage medium, the computer readable storage medium storing instructions readable by a processing circuit and executable by the processing circuit to perform a method for performing steps of the present invention.

[0006] Viewed from a further aspect, the present invention provides a computer program stored on a computer readable medium and loadable into an internal memory of a digital computer, the computer program comprising software code portions for performing the steps of the present invention when said program is run on a computer.

[0007] Viewed from a further aspect, the present invention provides a computer program product for implementing a method for searching images, the computer program product comprising a computer readable storage medium having program instructions executed by at least one computer processor to cause the computer processor to: identify a reference object in a two-dimensional reference image; determine a three-dimensional reference orientation axis of the reference object based on at least one attribute of the reference object; receive an input requesting a change in the three-dimensional perspective of the reference object; determine a new orientation axis based on the input and the reference orientation axis; perform a search query on the set of two-dimensional images based on the new orientation axis and the at least one attribute of the reference object; and display image search results ranked based on correlation with the new orientation axis and the at least one attribute of the reference object.

[0008] An embodiment of the present disclosure relates to a method for searching images. The method includes identifying an object in a two-dimensional reference image. The method includes determining a three-dimensional reference orientation axis of the object based on at least one attribute of the object. The method includes receiving an input requesting a change in the three-dimensional perspective of the object. The method further includes determining a new orientation axis based on the input and the reference orientation axis. The method includes performing a search query on the set of two-dimensional images based on the new orientation axis and the at least one attribute of the object. The method also includes displaying image search results ranked based on correlation with the new orientation axis and the attribute of the object.

[0009] Other embodiments of the present disclosure relate to computer systems and computer program products for the implementation of the methods.

[0010] The above summary is not intended to describe each illustrated embodiment or every implementation of the present disclosure.

[0011] The drawings included in this application are incorporated in and form a part of the specification. The drawings illustrate embodiments of the present disclosure and, together with the description, explain the principles of the disclosure. The drawings are only intended to illustrate certain embodiments and are not intended to limit the disclosure. [Brief description of the drawings]

[0012] [Figure 1] FIG. 2 is a block diagram of a processing system according to an embodiment. [Diagram 2] FIG. 1 is a block diagram of an illustrative cloud computing environment having one or more computing nodes with which local computing devices used by a cloud customer communicate, according to an embodiment. [Diagram 3] FIG. 1 is a block diagram of a set of functional abstraction layers provided by a cloud computing environment according to an embodiment. [Figure 4] 1 is a flowchart of an image search method according to an embodiment. [Figure 5A] FIG. 2 illustrates an example of an object shown in one of two different visual orientations, according to one embodiment. [Figure 5B] FIG. 2 illustrates an example of an object shown in one of two different visual orientations, according to one embodiment. [Figure 6] 1 illustrates an example of an object and a user's rotation gesture according to one embodiment. [Figure 7] FIG. 2 illustrates an example of a spherical model showing objects and viewpoints available in image search results, according to one embodiment. [Figure 8] FIG. 2 illustrates an example of a spherical model showing available viewpoints in image search results and the degree of correlation between available images and a reference image, according to one embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Image search engines do not provide 2D image results based on a particular viewing angle of an object in an image, nor do they allow users to specify a particular angle for viewing, however, it is desirable to be able to view certain objects as 2D images from different angles that do not necessarily lead to them being represented by a 3D image model.

[0014] The embodiments described herein provide systems, methods, and computer program products that allow a user to input a desired change in viewpoint of an object in a 2D image and receive 2D image search results that represent the desired viewpoint. Additionally, the embodiments described herein enable crowdsourcing of new images for missing viewpoints based on a user request to view a particular object from a particular viewpoint.

[0015] Referring now to the drawings, where like numerals represent the same or similar elements, and initially to Figure 1, an exemplary processing system 100 to which the present embodiments may be applied is shown according to one embodiment. Processing system 100 includes at least one processor (CPU) 104 operably coupled to other components via a system bus 102. Cache 106, read only memory (ROM) 108, random access memory (RAM) 110, input / output (I / O) adapter 120, audio adapter 130, network adapter 140, user interface adapter 150, and display adapter 160 are operably coupled to system bus 102.

[0016] The first storage device 122 and the second storage device 124 are I / O Operatively coupled to system bus 102 by adapter 120. Storage devices 122 and 124 may be either disk storage devices (e.g., magnetic or optical disk storage devices), solid state magnetic devices, etc. Storage devices 122 and 124 may be the same type of storage device or different types of storage devices.

[0017] The speakers 132 are operably coupled to the system bus 102 by an audio adapter 130. The transceiver 142 is operably coupled to the system bus 102 by a network adapter 140. The display device 162 is operably coupled to the system bus 102 by a display adapter 160.

[0018] A first user input device 152, a second user input device 154, and a third user input device 156 are operably coupled to the system bus 102 by a user interface adapter 150. The user input devices 152, 154, and 156 may be a keyboard, a mouse, a keypad, an image capture device, a motion detection device, a microphone, a device incorporating the functionality of at least two of the previous devices, or any other suitable type of input device. The user input devices 152, 154, and 156 may be the same type of user input device or different types of user input devices. The user input devices 152, 154, and 156 are used to input and output information to and from the system 100.

[0019] An image analysis component 172 is operatively coupled to the system bus 102. The image analysis component 172 (or engine) identifies objects in an image based on image analysis, image processing, metrology, edge detection, object detection, classification, etc., performed on the image. The image analysis component 172 is configured to identify a variety of different objects based on many different attributes of the objects. Object attributes may be identified based on color, model, object type, object shape, object size, etc. Furthermore, using associated image classification techniques, image objects are classified based on the identified attributes.

[0020] The image search engine component 174 is operatively coupled to the system bus 102. The image search engine component 174 searches for images based on keywords, pictures, web links to pictures, image metadata, color distribution, shape, rotational viewpoint, and the like.

[0021] Processing system 100 may include other elements (not shown) or omit certain elements, as will be readily contemplated by one of ordinary skill in the art. For example, various other input and / or output devices may be included within processing system 100 according to the same particular implementation, as will be readily understood by one of ordinary skill in the art. For example, various types of wireless and / or wired input and / or output devices may be used. Furthermore, additional processors, controllers, memories, etc. in various configurations may also be utilized, as will be readily appreciated by one of ordinary skill in the art. These and other variations of processing system 100 will be readily contemplated by one of ordinary skill in the art given the teachings of the disclosure provided herein.

[0022] Although this disclosure includes a detailed description of cloud computing, it should be understood that implementations of the teachings detailed herein are not limited to cloud computing environments. Rather, embodiments of the present disclosure may be implemented in connection with any other type of computing environment now known or later developed.

[0023] Cloud computing is a service delivery model designed to enable convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and services) that can be rapidly provisioned and released with minimal administration or interaction with the service provider.

[0024] A cloud computing environment is a service oriented approach with a focus on statelessness, loose coupling, modularity and semantic interoperability. At the heart of cloud computing is an infrastructure that includes a network of interconnected nodes.

[0025] Referring now to FIG. 2, an illustrative cloud computing environment 250 is depicted. As shown, the cloud computing environment 250 includes one or more cloud computing nodes 210 with which local computing devices used by cloud consumers, such as a personal digital assistant (PDA) or mobile phone 254A, a desktop computer 254B, a laptop computer 254C, or an automobile computer system 254N, or combinations thereof, may communicate. The nodes 210 may communicate with each other. They may be physically or virtually grouped in one or more networks, such as in a private, community, public, or hybrid cloud, or combinations thereof, as previously described (not shown). This enables the cloud computing environment 250 to provide infrastructure, platform, and / or software as a service without the cloud consumer having to maintain resources on the local computing device. It should be understood that the types of computing devices 254A-N shown in FIG. 2 are for illustrative purposes only, and that computing nodes 210 and cloud computing environment 250 may communicate with any type of computerized device via any type of network and / or network-addressable connections (e.g., using a web browser).

[0026] Referring now to Figure 3, a set of functional abstraction layers provided by cloud computing environment 250 (Figure 2) is shown. It should be understood in advance that the components, layers and functions shown in Figure 3 are for illustration purposes only, and embodiments of the present invention are not limited thereto. As depicted, the following layers and corresponding functions are provided:

[0027] Hardware and software layer 360 includes hardware and software components. Examples of hardware components include mainframe 361, RISC (reduced instruction set computer) architecture based server 362, server 363, blade server 364, storage device 365, and network and networking components 366. In some embodiments, software components include network application server software 367 and database software 368.

[0028] The virtual layer 370 provides an abstraction layer within which the following examples of virtual entities may be provided: virtual servers 371, virtual storage 372, virtual networks including virtual private networks 373, virtual applications and operating systems 374 and virtual clients 375.

[0029] In one example, the management layer 380 may provide the functions described below. Resource provisioning 381 provides dynamic procurement of computational resources and other resources utilized to perform tasks within the cloud computing environment. Metering and pricing 382 provides tracking of costs as resources are utilized within the cloud computing environment and provides for the creation or delivery of invoices for the consumption of these resources. In one example, these resources may include application software licenses. Security provides protection of data and other resources as well as identity verification for cloud consumers and tasks. User portal 383 provides consumers and system administrators with access to the cloud computing environment. Service level management 384 provides allocation and management of cloud computing resources such that required service levels are met. Service level agreement (SLA) planning and fulfillment 385 provides pre-arrangement and procurement for cloud computing resources where future demand is anticipated according to SLAs.

[0030] The workload layer 390 provides examples of functions for which a cloud computing environment may be utilized. Examples of workloads and functions provided from this layer include mapping and navigation 391, software development and lifecycle management 392, virtual classroom instructional delivery 393, data analytics processing 394, transaction processing 395, and image search engine processing 396.

[0031] In certain embodiments, the image search engine receives input from a user regarding a change in the angular perspective of a 2D object in an image and searches for existing images stored in storage that match this new perspective. Thus, there is a collection of different 2D images of an object stored at multiple different angular perspectives. This differs from mechanisms for changing the angular viewing perspective for a three-dimensional object (e.g., an object generated in a CAD program), where the system programmatically generates an entirely new image every time there is a change in the viewing angle and / or zoom level.

[0032] Referring now to FIG. 4, in a particular embodiment, in step 400, an image search engine identifies an object in an image. In step 402, the search engine determines an orientation vector of the identified object. In step 404, the search engine receives input from a user for rotation of the object in the image. In one example, the input from the user can be a rotation gesture of the user's hand, as described in more detail herein. Based on the input from the user and the magnitude of the change in the orientation vector, the image search engine searches the set of images to determine whether other 2D images of the object are available at the new orientation vector. In step 406, the search engine executes a search query based on the new orientation vector and based on one or more attributes of the image (e.g., the color of the object). If in step 408, the search engine determines that another 2D image is available, then in step 410, the new image is displayed to the user on the display device. If in step 408, the search engine determines that no 2D image is available for the new orientation vector, there will be no image displayed to the user. The search engine recognizes that there is missing information, and in step 412, the search engine requests the missing information from one or more sources. As described in more detail herein, in certain embodiments, the request for additional information can be a crowdsourcing request. With crowdsourcing, a large number of people are asked to provide images of themselves to the image search engine, and these provided images help complete the set of images so that the user can view the object from any 360 degree rotational perspective. After step 410 or step 412 is completed, the system returns to step 404 so that the user may enter another rotation gesture. Furthermore, in certain embodiments, in step 412, if the request for the missing image is fulfilled by an external source that provides such an image, the repository of the image may be updated with the new image (e.g., automatically or after approval of the image by a system administrator), and future search queries are performed based on the updated repository of the image.

[0033] In a particular embodiment, a user is presented with a 2D image that includes one or more objects. In step 400, an image search engine analyzes the 2D image and identifies objects present in the image. Identification of the objects is based on image analysis, image processing, metrology, edge detection, object detection, classification, etc. performed on the image. The image analysis engine can identify a variety of different objects based on many different attributes of the objects. Object attributes may be identified based on color, model, object type, object shape, object size, etc. Furthermore, using related image classification techniques, the image objects are classified based on the identified attributes.

[0034] After identifying the presence of an object, the search engine determines in step 402 which direction the object is facing. In particular, the search engine identifies a baseline direction (or orientation vector) for each object in the image. In one embodiment, the orientation vector is an axis that passes through the object's calculated center of mass, and this vector extends in a direction from the front of the object to the back of the object. This initial orientation vector of the object, together with the user's rotation gesture, is the reference used to calculate a new orientation vector. In one embodiment, the initial orientation vector of the object appearing in the 2D image forms part of the image's metadata and is used by the search engine when searching for new images. It should be understood that in certain embodiments, the orientation vector may not pass through the object's center of mass, may not extend from the front to the back, and may extend in a different direction relative to the object.

[0035] In certain embodiments, there is a library of different orientation directions for different classification types of objects. In these embodiments, the data in the library indicates a reference orientation for an animal (e.g., the front of a face), or a car (e.g., the front side of a car), or a house (e.g., the side of a building with a front door), or any other classification type. In these embodiments, once the reference side of an image object is determined, the image analysis module assigns a reference orientation axis to the image object. The assigned axis is the reference axis for the recognized object. Once the reference axis of any image object is identified, other image objects present in the same or different images are compared and a relative reference orientation or side is calculated. The relative object orientation of the different image object is compared to the reference image object and an orientation axis is calculated based on the difference. In one embodiment, the orientation of the different image object is calculated relative to the reference image object and assigned as the current angular orientation of the object. The angular orientation of the object is calculated relative to the 3D axis, and thus the angular orientation of the 3D axis is calculated and is part of the metadata of the image object.

[0036] 5A and 5B, in a particular embodiment, a 2D image (e.g., a digital image of a photograph) including at least one object (e.g., a car) is displayed on a display device. The image is a 2D image, and the object is displayed to the user from an initial three-dimensional angular viewpoint (i.e., perspective). With reference to FIG. 5A, there is a reference image viewpoint for the object 500. In the example shown in FIG. 5A, the reference image is a front view of the object 500, which may be on the Z and X axes. However, it should be understood that any other suitable viewpoint other than the front view may be used for the reference viewpoint. The reference viewpoint also has a corresponding reference orientation vector perpendicular to the other two axes. With reference to FIG. 5A, the X-axis 504 and the Z-axis 502 are shown extending horizontally and vertically, and the reference orientation is along the Y-axis 506 direction perpendicular to the other axes (see Y-axis 506 in FIG. 5B). In one embodiment, the Y-axis 506 is the object's orientation axis (or orientation vector 506), which runs through the object's center of mass from the front to the back of the object. With reference to Figure 5B, the object 500 has an orientation vector 506 that is angled forward and to the left and slightly downward. The direction of the orientation vector 506 in Figure 5B has a particular angular offset in each of the X-axis 504, Y-axis 506, and Z-axis 502 directions from the reference orientation vector in Figure 5A.

[0037] Referring now to FIG. 6, in one embodiment, if a user desires to view the object 600 from a different orientation or perspective, the user may request a different angular perspective by providing input to the image search engine. In response to the user's rotation gesture or other input (e.g., a rotation gesture 604 of the user's hand 602), the image search engine determines a desired amount of change in the rotational perspective of the object 600 in the 2D image and retrieves a different 2D image corresponding to the new perspective requested by the user. For example, as shown in FIG. 6, the user has performed a rotation gesture 604, requesting an angular change 606 in the perspective of the object 600. Referring again to FIG. 5B, the object 500 is presented to the user at different angular perspectives relative to the reference perspective shown in FIG. 5A or FIG. 6. The user may request a change to the angular perspective by providing gesture input any number of times. In a particular embodiment, if a different 2D image does not exist for the new rotational perspective, the search engine initiates a request to one or more resources to provide the missing image information.

[0038] The rotation gesture input that a user provides to a search engine for a change in rotational perspective can be configured to differ depending on the type of input device available to the user.

[0039] In one embodiment, if the input device is a mouse, the user can perform a click-and-drag action on the mouse to request a new perspective of the object. For example, the click-and-drag action can specify a distance and angle relative to an origin (e.g., the location where the mouse was clicked or a mouse down event). The distance and angle from the origin to the final mouse location (e.g., the location where the mouse was released or a mouse up event) is a vector that can be used to determine the desired rotation of the object. In another example, the object has a center location, and the distance between this center location and the locations of the mouse down event and the mouse up event are all used to determine the desired rotation of the object.

[0040] In one embodiment, if the input device is a touch screen enabled display device, the user can use one or more fingers to perform touch, drag and rotate actions to request an orientation change. With one finger, this rotation request can be performed similarly to the click and drag input of a mouse described above. With multiple fingers, the user can pinch or stretch the touch screen to further enable zoom in or zoom out actions.

[0041] In one embodiment, where the input device is an augmented reality (or virtual reality) display device, the user can perform a rotation gesture (e.g., by rotating their hand) to request a change in orientation. In these particular embodiments, the user can perform a 3D rotation gesture on a 2D object based on Internet of Things ("IoT") data or other camera output, and with this information the system can identify the rotation direction and angular motion of the gesture. For example, a camera can take a picture of a person's hand, and through image processing, changes in the position, size and orientation of the hand can be determined. Based on these changes, a rotation gesture is determined. It should be understood that any other suitable type of input may be provided by the user, provided that the input reflects the desired change in the angular perspective of the object.

[0042] In one embodiment, in response to a rotation gesture (or other suitable input) by the user, the search engine uses at least one of the direction of the rotation gesture, the angular motion of the rotation gesture, the magnitude of the rotation gesture, and the angular orientation (i.e., orientation vector) of the object currently displayed in the 2D image. Thus, the input indicates a change in the orientation vector relative to the current orientation vector. Based on this input data, the search engine calculates a new orientation vector for the object, and based on this data, the search engine constructs a new search query. The search engine executes the search query and returns one or more images of the object having an orientation vector that is consistent with (or at least close to) the calculated new orientation vector. This new image is displayed to the user on the display device. In this way, the user can understand how the object looks from different perspectives without having to resort to complex 3D object manipulation programs (e.g., CAD programs) simply by utilizing a collection of different 2D images. Furthermore, the object may not be suitable for viewing as a three-dimensional object in a CAD program (e.g., a mountain range, a tourist landmark, a house, a building, etc.). This process of selecting a new perspective can be repeated any number of times by the user to view the object in different ways.

[0043] In certain embodiments, the search engine uses one or more attributes of the object identified in the 2D image to perform the search query. As described above, the search engine performs image analysis of the object in the 2D image to determine the baseline orientation vector of the object. However, the image analysis can be used to identify one or more additional attributes of the object to assist the search query. Additionally, metadata of the image file can be used to assist the search query. Examples of other attributes include the size of the object, the color of the object, the type or category of the object, the geographic location where the image was captured, the timestamp when the image was captured, the product part number of the object, other metadata of the image file, such as image resolution, graphical features of the object, etc.

[0044] In one embodiment, in response to a rotation gesture input from the user, the search engine executes a search query and searches for new 2D images of objects with certain attributes that correlate to those of the original 2D image. For example, if a user is looking at a picture of a red sedan car from a first viewing angle and inputs a rotation gesture to look at a new perspective, if the search engine presents the user with a new 2D image of a blue pickup truck, this will not be useful to the user. Although they are both cars, they look quite different. If the new image does not at least substantially visually correspond to the original image, the user may feel a sense of disconnection with the original image. Based on these identified attributes, the search engine attempts to find new 2D images in new orientations that have similar (or identical) visual attributes to the original 2D image. Thus, while the viewer is not actually looking at different perspectives of the exact same 3D object, it should feel more natural to the user to see different 2D images with very similar visual characteristics.

[0045] In certain embodiments, the search engine requires that all 2D images are of the exact same object. For example, if a company has a website with an online product catalog, the company may want to ensure that different views of the object are not from slightly different products to avoid potential customer confusion when purchasing products. In one example, the product part number (or serial number) is an image metadata attribute of the different 2D images, which helps ensure that image search results do not return views of different products.

[0046] In certain embodiments, when a search engine displays 2D image search results, the search engine also displays an indication to the user as to what other perspectives are available. There may be situations where a search engine has located many stored images of an object from a front-facing perspective, but has few (or no) collections of images of the product from rear-facing views. The system presents the user with a visual indication of the availability of 2D images so that the user can make an informed decision about which rotation view to select next. For example, if the user knows that there are no rear-facing views of the 2D object in the collection, they will not waste time performing a rear-facing rotation gesture.

[0047] Referring now to FIG. 7, in an embodiment, a visual indication to a user of available viewpoints of an object 702 in an image search result is shown as a three-dimensional sphere 706. In this example, the sphere 706 has shaded portions indicating the presence of 2D images of the object 702 at different angle viewpoints. As shown in FIG. 7, the 3D wireframe representation of the sphere 706 has an orientation vector 704. Prior to input of a rotation gesture by the user, the display device displays a current viewpoint of the object 702 with an orientation vector 704 that is identical to the orientation vector 704 of the sphere 706. In a particular embodiment, the user may interact with the input device by a rotation gesture, and the image search engine causes a rotation of the sphere 706 on the display device. The new orientation vector 704 of the sphere 706 provides the user with a preview of what the new viewpoint of the object 702 will be. In a particular embodiment, the image search engine does not update the 2D image of the object 702 until the user completes the rotation gesture. In other embodiments, as the user continues to rotate the sphere 706, the image search engine continuously searches and updates images of the object 702 (if available).

[0048] In the example shown in FIG. 7, portions of the wire-frame sphere 706 are shaded in a binary manner (e.g., black portions 708 and white portions 710) to indicate which angular viewpoints have 2D images available. In this example, if the user provides a rotation gesture to the sphere 706, it will rotate to show a new front-most partial portion of the sphere. If this front-most spherical portion is shaded black, this will mean that an image search at this angle will not produce results. Thus, the user will know to continue the rotation gesture to different angles that will have image results. In one embodiment, if the user rotates the sphere 706 to an angle that does not have image results, the search images will find results at different angular viewpoints that most closely approximate the one requested by the user. It should be appreciated that it is not required that the front-most portion of the sphere be the portion of the new viewpoint, and any suitable portion can be used. In certain embodiments, the portion of the sphere 706 that corresponds to the new viewpoint to be selected is highlighted, colored, or in some other way indicated to the user.

[0049] In certain embodiments, the portions of the wireframe sphere are not limited to being presented to the user in a binary manner, and the portions have additional visual indications or signs that are displayed to the user. For example, if the original 2D image of the object is a blue sedan car, then some of the different viewpoints may have different 2D images of the blue sedan car. In this situation, the portions of the wireframe sphere that correspond to these viewpoints may be colored blue. However, for other angular viewpoints, only images of red sedan cars may be present. For these angular viewpoints, the portions of the wireframe sphere that correspond to these viewpoints may be colored red. This alerts the user that for these viewpoints, the user will still see a sedan car, but it will not be the same color as the color of the original 2D image of the object. It should be understood that the types of indications or signs on the sphere are not limited to different colors. The types of signs may represent some other visual attribute of the object (e.g., size, type, height, etc.).

[0050] With reference to FIG. 8, in certain embodiments, the type of indication may also be any suitable visual indication of how closely the object in different views matches the original 2D image of the object. In one embodiment, as shown in FIG. 8, a heat map of a sphere 800 is presented that represents the degree to which the object in different perspectives matches (or differs) from the original image of the object, even if all the different 3D perspectives have an existing 2D image of the object. With reference to FIG. 8, a sphere 800 is shown with an orientation vector 802. In this example, the sphere 800 has multiple surface portions. Some of the portions are white 804 (or clear), indicating that there is no existing image for that viewing perspective. Other portions are different shades of gray. A light gray portion 806 indicates that there is an existing image for that viewing perspective, but the visual characteristics of the object in that image do not match the original image very closely. The darkness of the shaded lines within the portions indicates how closely the existing image matches the original image. In this regard, the black portions 808 match the original image much more closely than the light gray portions (e.g., gray portion 806). Thus, this heat map of sphere 800 provides a user with an easy to understand map of where closely matching images exist and where they do not.

[0051] In certain embodiments, angular views that match very closely to the original image (e.g., same color and size) may be shown in one color on the 360-degree heat map, while other angular views that match less closely (e.g., meet the required angular criteria but differ in color and / or size from the original image) may be shown in a different color on the 360-degree heat map (e.g., shown as red on the heat map versus green). It should be appreciated that the sphere 800 (or other indicator) does not necessarily need to be divided into individual surface portions, and any other suitable visual indication may be used to represent the degree to which the heat map and different existing 2D images match the original image.

[0052] In other embodiments, the new angular orientation is determined based on the user's gesture and the angular orientation of the original image object, and the image search results are ranked based on their deviation from the new angular orientation. When the search results are displayed, the image search engine analyzes the attributes of the image object and identifies the availability of other images from different sources.

[0053] It should be understood that the viewpoint indicator need not be a 3D wireframe sphere. It can be any other suitable object or visual indication that allows the user to select different three-dimensional viewpoints by a rotation gesture (or other input type). Thus, if the object is a cow, the visual indication can be a 3D wireframe representation of a cow or any other appropriately formed object. Furthermore, the visual indication need not be a 3D wireframe model. The visual indication can be any other suitable three-dimensional shape with different surface indicia that indicate the amount (or absence) of image search results. For example, the visual indication can be a smooth sphere with a heat map (or color gradient) that indicates the availability of search results at different angles. In other embodiments, the visual indication is a 2D representation rather than a three-dimensional model.

[0054] In certain embodiments, the search engine implements a method for identifying, requesting, and replenishing missing image information (or requesting better image information). There may be situations where a complete set of 2D images does not exist for every possible angular viewpoint for a given object. In response to a rotation gesture and an image search query that does not return results, the search engine initiates a request to one or more recipients for additional image data. In one embodiment, the request to one or more recipients includes a crowdsourcing image acquisition effort. In general, crowdsourcing refers to a sourcing model in which an individual or organization acquires goods and services, including ideas and financial resources, from a large and dynamic group of providers (e.g., Internet users). Thus, crowdsourcing shares an effort among many participants to achieve a cumulative result. In the context of this embodiment, the goal is to acquire a very large number of relevant images for a wide variety of different objects and in a large number of different orientations for the objects. Thus, there may be a very large effort required to create such a large collection of 2D images, and crowdsourcing can be used to mitigate or reduce the magnitude of this effort. For example, if the target object is a tourist attraction such as the Eifel Tower, tourists may willingly submit their vacation photos of the tower from many different angles. This helps to create a highly relevant set of 2D images of the tower from various perspectives. In certain embodiments, the crowdsourced contributors are the general public (e.g., all Internet users). In other embodiments, the crowdsourced contributors are a limited group of users, e.g., employees of a company. In one embodiment, the original content provider offers contributors of other viewing angles of the parent image compensation (e.g., monetary compensation, discounts on products / services, etc.) for providing their images.

[0055] In one embodiment, the trigger for creating a crowdsourcing request for an image is when a user performs a rotation gesture to view an object from a particular viewpoint (with the additional requirement that there is no existing image for that viewpoint). As described above with respect to FIG. 4, if it is determined in step 406 that there is no image for the new rotation vector, then the image search engine sends a request for the missing image information in step 410. In other words, if a user performs a request to view an object from a particular viewpoint, it indicates that there is an actual demand to view the object from that angle. However, if there has never been a request to view a product from the back, for example, accumulating a photo of that object from the back through a crowdsourcing effort may not be worth the effort. In an example where the object is a car, there may be little or no demand to view the underside of the car. In this example, people are not interested in viewing the object from that angle.

[0056] In one embodiment, when a donor responds to a request by providing additional images, the owner of the parent image (i.e., the reference image) has the option to review the images. The owner of the reference image may review the provided images and determine whether they are of acceptable quality or match the visual attributes of the parent image to a sufficient degree. If the owner of the parent image decides that the provided images are acceptable, they may store the images in storage and make them publicly available for viewing additional viewing angles of the images. In other embodiments, there may be no owner of the parent image and no one to manually review the provided images. In this embodiment, image processing is performed on the provided image to determine whether it is an appropriate match to the parent image.

[0057] In one embodiment, even if an image search query produces some results, if the matching quality of the image for that rotation viewer is low, the image search engine may still initiate a request for additional image data. For example, there may be an existing 2D image of an object for a particular angle perspective that has a very low correlation to the object in the original reference image. In one example, the original reference image may be a gray short bed pickup truck, and the stored image for the other perspective is a brown long bed pickup truck with a different make and model than the original. In this example, perhaps the brown long bed pickup truck had some visual attributes that met some threshold that would be considered an acceptable match. However, this brown truck has a very low match with the original image. In this embodiment, in response to a determination that the stored image for the rotation gesture and the new orientation vector has a low correlation with the reference image (i.e., is below a certain correlation threshold), the search engine sends a request to one or more recipients to update the image and improve the quality of the match. Upon receiving a new image from the provider in response to the request, the system compares the newly received image with the previous image. If the new image is determined to be a better match (i.e., better correlates with attributes of the original reference image than the current image), the search image replaces the current image with the new image. In one example, when the reference orientation axis of the received image and at least one attribute of the received image object correspond to a predetermined degree (e.g., this can be set by the content provider) with the reference orientation axis and at least one attribute of the reference image, the system adds the received image to the set of two-dimensional images. In this manner, by accepting new images and potentially replacing old images, the 360-degree rotational collection of 2D images for a given object can be continuously improved over time.

[0058] In one embodiment, the system utilizes dynamic contextually relevant context re-rendering based on object session usage type, task, purpose, logical progression of objects related to: local user historical machine learning, crowd-based usage for various unique users (sharing context), crowd-sourced user machine learning based on object orientation, and user rotation object flapping predictions.

[0059] In one embodiment, the system utilizes physical location refinement by crowdsourcing to geographically pinpoint people who are most likely to assist with providing content to complete a model or objective. For example, if there are geographically fenced areas, the system selectively looks at each person's current dynamic location and makes a determination as to whether their unique perspective is beneficial for the perspective and time (i.e., temporal decision). This method has the ability to selectively include only certain people within specific marked locations to target temporal events or schedules, thus deriving a unique and more complete model.

[0060] The present invention may be a system, method, or computer program product, or combination thereof, at any possible level of technical detail integration. The computer program product may include a computer readable storage medium having computer readable program instructions for causing a processor to perform aspects of the present invention.

[0061] A computer readable storage medium may be any tangible device capable of holding and storing instructions for use by an instruction execution device. A computer readable storage medium may be, for example, without limitation, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of computer readable storage media includes portable computer diskettes, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disk read only memory (CD-ROM), digital versatile disk (DVD), memory sticks, floppy disks, punch cards or mechanically encoded devices such as ridge structures in grooves with instructions recorded thereon, and any suitable combination of the foregoing. A computer-readable storage medium as used herein should not be construed as a signal that is itself ephemeral, such as an electric wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse passing through a fiber optic cable) or an electrical signal transmitted through a wire.

[0062] The computer readable program instructions described herein can be downloaded from the computer readable storage medium to each computing / processing device or to an external computer or storage device via a network, such as the Internet, a local area network, a wide area network, or a wireless network, or a combination thereof. The network may include copper transmission cables, optical transmission fiber, wireless transmission, routers, firewalls, switches, gateway computers, or edge servers, or a combination thereof. A network adapter card or network interface of each computing / processing device receives the computer readable program instructions from the network and transfers the computer readable program instructions to a computer readable storage medium within the respective computing / processing device for storage.

[0063] The computer readable program instructions for carrying out the operations of the present invention may be either source code or object code written in any combination of one or more programming languages, including assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state setting data, configuration data for integrated circuits, or object oriented programming languages ​​such as Smalltalk®, C++, and procedural programming languages ​​such as the “C” programming language or similar programming languages. The computer readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may execute computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry to perform aspects of the present invention.

[0064] Aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer readable program instructions.

[0065] These computer readable program instructions may be provided to a processor of a computer or other programmable data processing apparatus such that the instructions, executed by the processor of the computer or other programmable data processing apparatus, create means for performing the functions / operations specified in the blocks of the flowcharts and / or block diagrams to produce a machine. These computer readable program instructions may be stored in a computer readable storage medium such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions implementing aspects of the functions / operations specified in the blocks of the flowcharts and / or block diagrams, and may direct a computer, programmable data processing apparatus or other device, or combination thereof, to function in a particular manner.

[0066] The computer readable program instructions may also be loaded into a computer, other programmable data processing device or other device to cause a sequence of operational steps to be performed on the computer, other programmable device or other device to generate a computer-implemented process such that the instructions, which execute on the computer, other programmable device or other device, perform the functions / operations specified in the flowchart and / or block diagram blocks.

[0067] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowcharts or block diagrams may represent a module, segment, or part of instructions, including one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions described in the blocks may be performed out of the order described in the figures. For example, two blocks shown in succession may in fact be accomplished as one step, or may be executed simultaneously or substantially simultaneously, with partial or total overlap in time, or the blocks may sometimes be executed in reverse order, depending on the functionality involved. It should also be noted that each block of the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified functions or acts or executes a combination of dedicated hardware and computer instructions.

[0068] The description of the various embodiments is presented for illustrative purposes and is not intended to be comprehensive or limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope of the described embodiments. The terms used in this specification are selected to best explain the principles of the embodiments, practical applications, or technical improvements over the technology found in the market, or to enable those skilled in the art to understand the embodiments disclosed in this specification.

Claims

1. 1. A method for retrieving an image, comprising: Identifying a reference object in a two-dimensional reference image; determining a three-dimensional reference orientation axis of the reference object based on at least one attribute of the reference object; receiving an input requesting a change in a three-dimensional viewpoint of the reference object; determining a new orientation axis based on the input and the reference orientation axis; performing a search query on a set of two-dimensional images based on the new orientation axis and the at least one attribute of the reference object; displaying image search results based on a correlation between the two-dimensional images obtained by the search query and the at least one attribute of the reference object; A method comprising:

2. The input includes information regarding a rotation gesture performed by a user. The method of claim 1.

3. The information includes a direction of the rotation gesture and an angular motion of the rotation gesture. The method of claim 2.

4. When the image search result does not produce any results, the method further includes sending a request to an external image provider resource to provide an image corresponding to the new orientation axis and the at least one attribute of the reference object.

4. The method according to claim 1 , wherein the first and second substrates are fused to the first substrate.

5. The method comprises: receiving an image from the external image provider resource; identifying a received image object within the received image; determining a received image reference orientation axis of the received image object; adding the received image to the set of two-dimensional images when the received image reference orientation axis and at least one attribute of the received image object correspond to a predetermined degree to the reference orientation axis and the at least one attribute of the reference image; Further comprising: The method according to claim 4.

6. the external image provider resource is a crowdsourcing source; The method according to claim 4 or 5.

7. the reference orientation axis is an axis that passes through the calculated center of mass of the reference object and extends from the front-facing side of the reference object to the rear-facing side of the reference object; 7. The method according to any one of claims 1 to 6.

8. The method comprises: Displaying a three-dimensional perspective of the sphere to a user; updating a rotational visible perspective of the sphere to track an input gesture of the user; Further comprising:

8. The method according to any one of claims 1 to 7.

9. the sphere includes a plurality of surface portions corresponding to a plurality of different viewing angles of the reference object, the surface portions having visual attributes indicative of whether an associated image of the reference object is present within the set of two-dimensional images for the corresponding viewing angles; The method according to claim 8.

10. the visual attributes of the surface portion further indicate a degree of correspondence between the relevant image of the set of images and the reference image.

10. The method of claim 9.

11. 1. A system for retrieving images, comprising: a computer readable storage medium having program instructions stored thereon; one or more processors configured to execute the program instructions to perform the method; The method comprises: Identifying a reference object in a two-dimensional reference image; determining a three-dimensional reference orientation axis of the reference object based on at least one attribute of the reference object; receiving an input requesting a change in a three-dimensional viewpoint of the reference object; determining a new orientation axis based on the input and the reference orientation axis; performing a search query on a set of two-dimensional images based on the new orientation axis and the at least one attribute of the reference object; displaying image search results based on a correlation between the two-dimensional images obtained by the search query and the at least one attribute of the reference object; Including, the system.

12. The input includes information regarding a rotation gesture performed by a user. The system of claim 11.

13. The information includes a direction of the rotation gesture and an angular motion of the rotation gesture. The system of claim 12.

14. When the image search result does not produce any results, the method further includes sending a request to an external image provider resource to provide an image corresponding to the new orientation axis and the at least one attribute of the reference object. A system according to any one of claims 11 to 13.

15. The system comprises: receiving an image from the external image provider resource; identifying a received image object within the received image; determining a received image reference orientation axis of the received image object; adding the received image to the set of two-dimensional images when the received image reference orientation axis and at least one attribute of the received image object correspond to a predetermined degree to the reference orientation axis and at least one attribute of the reference image; Further comprising: The system of claim 14.

16. A computer program product that causes a computer to carry out the method according to any one of claims 1 to 10.

17. 17. A computer readable storage medium having the computer program of claim 16 stored therein.

Citation Information

Patent Citations

  • Image information generated from three-dimensional image information and image display data reproducing device

    JP1994020030A

  • Image display device, image display method, and program for displaying three-dimensional shape as two-dimensional image

    JP2011192242A

  • Image operation apparatus, image operation method and program

    JP2012208705A

  • Information display device, distribution device, information display method and information display program

    JP2016091239A

  • Imaging apparatus, information processing system, information processor, and polarization image processing method

    JP2018061124A