Google Patent | Digital Supplement Association And Retrieval For Visual Search

Patent: Digital Supplement Association And Retrieval For Visual Search

Publication Number: 20200159377

Publication Date: 20200521

Applicants: Google

Abstract

Systems and methods for identification and retrieval of content for visual search are provided. An example method includes transmitting a visual-content query to a server computing device and receiving a response to the visual-content query that identifies a digital supplement. The example method also includes causing a user interface to be displayed that includes information associated with the digital supplement. The visual-content query may be based on an image. The digital supplement may include information about the content of the image.

CROSS-REFERENCE TO RELATED APPLICATION

[0001] This application is a continuation application of U.S. application Ser. No. 16/014,512, filed on Jun. 21, 2018, the entirety of which is incorporated herein by reference.

BACKGROUND

[0002] Mobile computing devices, such as smartphones, often include cameras. These cameras can be used to capture images of entities in the environment around the computing device. Various types of content or experiences that relate to those entities may be available for users via the mobile computing device.

SUMMARY

[0003] This disclosure describes systems and methods for digital supplement association and retrieval for visual search. For example, systems and techniques described herein may be used to provide digital supplements, such as augmented reality (AR) content or experiences, that are responsive to a visual search. The visual search may for example be based on an image or an entity identified within an image. The digital supplement may, for example, include providing information or functionality associated with the image.

[0004] One aspect is a method that includes transmitting a visual-content query to a server computing device and receiving a response to the visual-content query that identifies a digital supplement. The method also includes causing a user interface to be displayed that includes information associated with the digital supplement.

[0005] Another aspect is a computing device comprising at least one processor and memory storing instructions. The instructions, when executed by the at least one processor, cause the computing device to capture an image and transmit a visual-content query that is based on the image to a server computing device. The instructions, when executed by the at least one processor, also cause the computing device to receive a response to the visual-content query that identifies a digital supplement and cause a user interface to be displayed that includes information associated with the digital supplement.

[0006] Yet another aspect is a computer-implemented method comprising capturing an image and transmitting a visual-content query based on the image to a search server. The method also includes receiving a response to the visual-content query that identifies a digital supplement server and causing the image to be transmitted to the digital supplement server. Additionally, the method includes receiving digital supplement content from the digital supplement server and causing the digital supplement content to be displayed.

[0007] The details of one or more implementations are set forth in the accompanying drawings and the description below. Other features will be apparent from the description and drawings, and from the claims.

BRIEF DESCRIPTION OF THE DRAWINGS

[0008] FIG. 1 is a block diagram illustrating a system according to an example implementation.

[0009] FIG. 2 is a third person view of an example physical space in which an embodiment of the client computing device of FIG. 1 is accessing digital supplements.

[0010] FIG. 3 is a diagram of an example method of enabling triggering of a digital supplement, in accordance with implementations described herein.

[0011] FIG. 4 is a diagram of an example method of enabling triggering of a digital supplement, in accordance with implementations described herein.

[0012] FIG. 5 is a diagram of an example method of searching for and presenting a digital supplement, in accordance with implementations described herein.

[0013] FIG. 6 is a diagram of an example method of identifying and presenting a digital supplement based on an image, in accordance with implementations described herein.

[0014] FIGS. 7A-7C are schematic diagrams of user interface screens displayed by embodiments of the client computing device of FIG. 1 to conduct a visual-content search and displaying a digital supplement.

[0015] FIGS. 8A-8C are schematic diagrams of user interface screens displayed by embodiments of the client computing device of FIG. 1 to conduct a visual-content search and displaying a digital supplement.

[0016] FIGS. 9A and 9B are schematic diagrams of user interface screens displayed by embodiments of the client computing device of FIG. 1 to conduct a visual-content search and display a digital supplement.

[0017] FIGS. 10A-10C are schematic diagrams of user interface screens displayed by embodiments of the client computing device of FIG. 1 to conduct a visual-content search and display a digital supplement.

[0018] FIGS. 11A-11C are schematic diagrams of user interface screens displayed by embodiments of the client computing device of FIG. 1 to conduct various visual-content searches within a store.

[0019] FIGS. 12A-12C are schematic diagrams of user interface screens displayed by embodiments of the client computing device of FIG. 1 during various visual-content searches.

[0020] FIG. 13 is a schematic diagram of an example of a computer device and a mobile computer device that can be used to implement the techniques described herein.

[0021] Reference will now be made in detail to non-limiting examples of this disclosure, examples of which are illustrated in the accompanying drawings. The examples are described below by referring to the drawings, wherein like reference numerals refer to like elements. When like reference numerals are shown, corresponding description(s) are not repeated and the interested reader is referred to the previously discussed figure(s) for a description of the like element(s).

DETAILED DESCRIPTION

[0022] The present disclosure describes technological improvements that simplify the identification and presentation of digital supplements based on visual content. Some implementations of technology described herein generate an index of digital supplements that are relevant to particular types of visual content and provide those digital supplements in response to a visual-content query received from a client computing device. This index can allow a user to access relevant digital supplements that are provided by network-accessible resources (e.g., web pages) disposed throughout the world.

[0023] For example, a client computing device, such as a smartphone, may capture an image of a supplement anchor, such as an entity. The client computing device may then transmit a visual-content query based on the image to a server computing device to retrieve digital supplements associated with the identified supplement anchor. In some implementations, the supplement anchor is based on the physical environment around the client computing device and the digital supplement is virtual content that may supplement a user’s experience in the physical environment.

[0024] The visual-content query may include the image or data that is determined from the image (e.g., such as an indicator of the identified supplement anchor). An example of data determined from the image is text that is extracted from the image using, for example, optical character recognition. Other examples of data extracted from the image include values read from barcodes, QR codes, etc., in the image, identifiers or descriptions of entities, products, or entity types identified in the image.

[0025] The entities, products, or entity types may be identified in the image using, for example, a neural network system such as a convolutional neural network system. The identifiers or descriptions of entities, products, or entity types may include metadata or a reference to a record in a database that relates to an entity, product, or entity type. Non-limiting examples of the entities include buildings, works of art, products, books, posters, photographs, catalogs, signs, documents (e.g., business cards, receipts, coupons, catalogs), people, and body parts.

[0026] Various types of digital supplements may be available that are related to a supplement anchor. The digital supplement may be provided by a network-accessible resource, such as a web page that is available on the Internet. There is a need for a way to locate and provide these digital supplements in response to a visual-content query. Some implementations generate and maintain an index of digital supplements that are associated with entities for use in responding to visual content queries. The index may, for example, be populated by crawling network-accessible resources to determine whether the network-accessible resources include or provide any are digital supplements and to determine the supplement anchors associated with those digital supplements.

[0027] For example, the network-accessible resource may include metadata that identifies the supplement anchors (e.g., text, codes, entities, or types of entities) for which a digital supplement is associated. The metadata may be included by the network-accessible resource in response to a hypertext transfer protocol (HTTP) request. The metadata may be provided in various formats such as extensible markup language (XML), JavaScript Object Notation (JSON), or another format.

[0028] The metadata for a digital supplement may include one or more of the following: a type indicator, an anchor indicator, a name, a description, a snippet of the content (i.e., an excerpt or preview of a portion of the content), an associated image, a link such as a URL to the digital supplement, and an identifier of an application associated with the digital supplement. The metadata may also include information about a publisher of the digital supplement. For example, the metadata may include one or more of a publisher name, a publisher description, and an image or icon associated with the publisher. In some implementations, the metadata includes context information related to providing the digital supplement. For example, the metadata may also include conditions (e.g., geographic conditions, required applications) associated with providing or accessing the digital supplement.

[0029] The identified digital supplements may be added to an index that is stored in a memory. In at least some implementations, the associated supplement anchor for a digital supplement is used as a key to the index. The digital supplements may also be associated with various scores. For example, a digital supplement may be associated with a prestige score that is based on how many other links are found (e.g., while crawling network-accessible resources) that reference the digital supplement or the network-accessible resource associated with the digital supplement and the prestige of the network-accessible resources that provide those links. As another example, a digital supplement may be associated with one or more relevance scores that correspond to the relevance of the digital supplement (or the associated network-accessible resource) to a particular anchor. The relevance score may also be associated with a keyword or subject matter. The relevance score may be determined based on one or more of the content of the digital supplement, the content of the network-accessible resource, the content of sites that link to the network-accessible resource, and the contents (e.g., text) of links to the network-accessible resources.

[0030] FIG. 1 is a block diagram illustrating a system 100 according to an example implementation. The system 100 may associate digital supplement with entities or entity types and may retrieve digital supplements in response to visual searches. A visual search is a search based on visual-content. For example, a visual search may be performed based on a visual-content query. A visual-content query is a query based on an image or other visual-content. For example, a visual-content query may include an image. In some implementations, a visual-content query may include text or data that is based on an image. For example, the text or data may be generated by recognizing one or more entities in an image. Some visual-content queries do not include an image (e.g., a visual-content query may include only data or text generated from an image). In some implementations, the system 100 includes a client computing device 102, a search server 152, and a digital supplement server 172. Also shown is a network 190 over which the client computing device 102, the search server 152, and the digital supplement server 172 may communicate.

[0031] The client computing device 102 may include a processor assembly 104, a communication module 106, a sensor system 110, and a memory 120. The sensor system 110 may include various sensors, such as a camera assembly 112, an inertial motion unit (IMU) 114, and a global positioning system (GPS) receiver 116. Implementations of the sensor system 110 may also include other sensors, including, for example, a light sensor, an audio sensor, an image sensor, a distance and/or proximity sensor, a contact sensor such as a capacitive sensor, a timer, and/or other sensors and/or different combinations of sensors. In some implementations, the client computing device 102 is a mobile device (e.g., a smartphone).

[0032] The camera assembly 112 captures images or videos of the physical space around the client computing device 102. The camera assembly 112 may include one or more cameras. The camera assembly 112 may also include an infrared camera. Image captured with the camera assembly 112 may be used to identify to supplement anchors and to form visual content queries.

[0033] In some implementations, images captured with the camera assembly 112 may also be used to determine a location and orientation of the client computing device 102 within a physical space, such as an interior space, based on a representation of that physical space that is received from the memory 120 or an external computing device. In some implementations, the representation of a physical space may include visual features of the physical space (e.g., features extracted from images of the physical space). The representation may also include location-determination data associated with those features that can be used by a visual positioning system to determine location and/or position within the physical space based on one or more images of the physical space. The representation may also include a three-dimensional model of at least some structures within the physical space. In some implementations, the representation does not include three-dimensional models of the physical space.

[0034] The IMU 114 may detect motion, movement, and/or acceleration of the client computing device. The IMU 114 may include various different types of sensors such as, for example, an accelerometer, a gyroscope, a magnetometer, and other such sensors. An orientation of the client computing device 102 may be detected and tracked based on data provided by the IMU 114 or GPS receiver 116.

[0035] The GPS receiver 116 may receive signals emitted by GPS satellites. The signals include a time and position of the satellite. Based on receiving signals from several satellites (e.g., at least four), the GPS receiver 116 may determine a global position of the client computing device 102.

[0036] The memory 120 may include an application 122, other applications 140, and a device positioning system 142. The other applications 140 include any other applications that are installed or otherwise available for execution on the client computing device 102. In some implementations, the application 122 may cause one of the other applications 140 to be launched to provide a digital supplement. In some implementations, some digital supplements may only be available if the other applications 140 include a specific application associated with or required to provide the digital supplement.

[0037] The device positioning system 142 determines a position of the client computing device 102. The device positioning system 142 may use the sensor system 110 to determine a location and orientation of the client computing device 102 globally or within a physical space. In some implementations, the device positioning system 142 determines a location of the client computing device 102 based on, for example, a cellular triangulation.

[0038] In some implementations, the client computing device 102 may include a visual positioning system that compares images captured by the camera assembly 112 (or features extracted from those images) to a known arrangement of features within the representation of the physical space to determine the six degree-of-freedom pose (e.g., the location and orientation) of the client computing device 102 within a physical space.

[0039] The application 122 may include a supplement anchor identification engine 124, a digital supplement retrieval engine 126, a digital supplement presentation engine 128, and a user interface engine 130. Some implementations of the application 122 may include fewer, additional, or other components.

[0040] The supplement anchor identification engine 124 identifies supplement anchors based on, for example, images captured with the camera assembly 112. In some implementations, the supplement anchor identification engine 124 analyzes an image to identify text. The text may then be used to identify an anchor. For example, the text may be mapped to a node in a knowledge graph. For example, the text may be recognized as the name of an entity such as a person, place, product, building, artwork, movie, or other type of entity. In some implementations, the text may be recognized as a phrase that is commonly associated with a specific entity or as a phrase that describes a specific entity. For example, the text may then be recognized as an anchor associated with the specific entity.

[0041] In some implementations, the supplement anchor identification engine 124 identifies one or more codes, such as a barcode, QR code, or another type of code, within an image. The code may then be mapped to a supplement anchor.

[0042] The supplement anchor identification engine 124 may include a machine learning module that can recognize at least some types of entities within an image. For example, the machine learning module may include a neural network system. Neural networks are computational models used in machine learning and made up of nodes organized in layers with weighted connections. Training a neural network uses training examples, each example being an input and a desired output, to determine, over a series of iterative rounds, weight values for the connections between layers that increase the likelihood of the neural network providing the desired output for a given input. During each training round, the weights are adjusted to address incorrect output values. Once trained, the neural network can be used to predict an output based on provided input.

更多阅读推荐......