Computer-implemented method and computing system for providing navigation instructions
Through computing systems and machine learning models, analyzing location data and semantic labels in geographic areas, determining the appropriate meeting locations between carriers and passengers, solving the problem of insufficient efficiency and safety in the existing technology, and achieving more efficient Navigation and resource utilization.
Patent Information
- Application Number
- CN201980052658.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2018-12-17
- Filing Date
- 2019-01-07
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2039-01-07
AI Technical Summary
The prior art is difficult to effectively determine the appropriate meeting location between carriers and passengers in a geographical area, especially under different movement forms and environmental conditions, affecting the efficiency and safety of navigation and meeting coordination.
The computing system receives location data and semantic tags, uses machine learning models to analyze image content, identify locations that meet meeting standards, generate navigation guidelines and instructions, and determine the meeting locations of carriers and passengers.
Improves the efficiency and security of meeting location determination, reduces resource consumption and waste of communication network bandwidth, provides more accurate and efficient navigation guidance, and enhances passenger and carrier safety.
Smart Images

Figure CN112543949B_ABST
Abstract
Description
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This application is based upon and claims the benefit of U.S. Provisional Patent Application No. 62 / 780,695, filed on December 17, 2018, which is incorporated herein by reference. Technical Field
[0003] The present disclosure relates generally to navigation and image content analysis. More specifically, the present disclosure relates to determining a meeting location using a computing device. Background Art
[0004] Operations associated with the state of a geographic area may be implemented on a variety of computing devices. These operations may include processing data associated with the geographic area for later access and use by a user or computing system. In addition, these operations may include exchanging data with remote computing systems. However, the types of operations performed and the purposes for which these operations are applied may change over time, as may the underlying hardware that implements these operations. Therefore, there are a variety of ways to utilize computing resources associated with determining the state of a geographic area. Summary of the invention
[0005] Aspects and advantages of embodiments of the present disclosure will be set forth in part in the following description, or may be learned from the description, or may be learned through practice of the embodiments.
[0006] An example aspect of the present disclosure is directed to a computer-implemented method for providing navigation instructions. The computer-implemented method may include accessing location data and multiple semantic tags by a computing system including one or more processors. The location data may include information associated with multiple locations, including the location of the carrier and the location of the passenger. The multiple semantic tags may include information associated with one or more features of the multiple locations. The computer-implemented method may also include identifying, by the computing system, location data and multiple semantic tags that satisfy one or more meeting criteria associated with the suitability of the location for the carrier and the passenger. The computer-implemented method may include determining, by the computing system, one or more meeting locations for the carrier and the passenger based at least in part on the location data and multiple semantic tags that satisfy the one or more meeting criteria. In addition, the computer-implemented method may include generating, by the computing system, one or more indications associated with the at least one meeting location based at least in part on one or more features of at least one of the one or more meeting locations.
[0007] Another exemplary aspect of the present disclosure is directed to one or more tangible, non-transitory computer-readable media storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform operations. These operations may include accessing location data and multiple semantic tags. The location data may include information associated with multiple locations, including the location of the carrier and the location of the passenger. The multiple semantic tags may include information associated with one or more features of the multiple locations. The operations may also include identifying location data and multiple semantic tags that meet one or more meeting criteria associated with the suitability of the location for the carrier and the passenger. The operations may include determining one or more meeting locations for the carrier and the passenger based at least in part on the location data and multiple semantic tags that meet the one or more meeting criteria. In addition, the operations may include generating one or more indications associated with the at least one meeting location based at least in part on one or more features of at least one of the one or more meeting locations.
[0008] Another example aspect of the present disclosure is directed to a computing system comprising one or more processors; and, one or more non-transitory computer-readable media storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations. The operations may include accessing location data and a plurality of semantic tags. The location data may include information associated with a plurality of locations, including locations of carriers and locations of passengers. The plurality of semantic tags may include information associated with one or more features of the plurality of locations. The operations may also include identifying location data and a plurality of semantic tags that satisfy one or more meeting criteria associated with the suitability of the location for the carrier and the passenger. The operations may include determining one or more meeting locations for the carrier and the passenger based at least in part on the location data and the plurality of semantic tags that satisfy the one or more meeting criteria. In addition, the operations may include generating one or more indications associated with the at least one meeting location based at least in part on one or more features of at least one of the one or more meeting locations.
[0009] Other aspects of the present disclosure are directed to various systems, devices, non-transitory computer-readable media, user interfaces, and electronic devices. These and other features, aspects, and advantages of various embodiments of the present disclosure will be better understood with reference to the following description and appended claims. The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate example embodiments of the present disclosure and, together with the description, are used to explain the relevant principles. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] A detailed discussion of the embodiments for those of ordinary skill in the art is set forth in the specification with reference to the accompanying drawings, in which:
[0011] Figure 1 A diagram depicting an example system according to an example embodiment of the present disclosure;
[0012] Figure 2 A diagram depicting an example device according to an example embodiment of the present disclosure;
[0013] Figure 3 An example of determining a meeting location according to an example embodiment of the present disclosure is depicted;
[0014] Figure 4 An example of determining a meeting location according to an example embodiment of the present disclosure is depicted;
[0015] Figure 5 An example of determining a meeting location according to an example embodiment of the present disclosure is depicted;
[0016] Figure 6 An example of determining a meeting location according to an example embodiment of the present disclosure is depicted;
[0017] Figure 7 An example of determining a meeting location according to an example embodiment of the present disclosure is depicted;
[0018] Figure 8 A flowchart depicting an example method for determining a meeting location in an environment according to an example embodiment of the present disclosure;
[0019] Fig. 9 A flowchart depicting an example method for determining a meeting location in an environment according to an example embodiment of the present disclosure;
[0020] Fig.10 A flowchart depicting an example method for determining a meeting location in an environment according to an example embodiment of the present disclosure;
[0021] Fig.11 A flowchart depicting an example method for determining a meeting location in an environment according to an example embodiment of the present disclosure;
[0022] Fig.12 A flowchart depicting an example method for determining a meeting location in an environment according to an example embodiment of the present disclosure;
[0023] Fig.13 A flowchart depicting an example method for determining a meeting location in an environment according to an example embodiment of the present disclosure; and
[0024] Fig.14 Depicted is a flow chart of an example method for determining a meeting location in an environment according to an example embodiment of the present disclosure.
[0025] Reference numerals repeated across the various drawings are intended to identify like features in the various embodiments. DETAILED DESCRIPTION
[0026] In general, the present disclosure relates to determining a meeting location for navigation or coordination of a meeting between parties. The present disclosure also describes generating an indication associated with a meeting location that satisfies meeting criteria associated with the suitability of the location for the parties (which may include carriers and passengers) based on features of the meeting location. In addition, the parties may have different mobility patterns, which may affect the visibility and / or accessibility of a given location to the parties (e.g., one party may be traveling by car, while the other party may be traveling on foot). Specifically, the disclosed technology may include a computing system that receives data including location data and semantic tags associated with features of an image (e.g., a photograph of an area captured from various vantage points). The location data may include a map of a geographic area and the locations of a carrier (e.g., a driver of a vehicle) and a passenger (e.g., a passenger intending to meet the carrier), and the semantic tags may be based on images that have been processed by a machine learning model or other image content analysis system configured to detect and identify various features of an image. Additionally, semantic tags may be associated with a geographic region, including specific locations within the geographic region (eg, latitude, longitude, and altitude) and other information (eg, amenities at each location).
[0027] The computing system may then identify location data and semantic tags that include information that satisfies meeting criteria associated with the suitability of the meeting location for the carrier and the passenger (e.g., a safe, easily visible, and convenient location that is not too far away). The computing system may also determine a meeting location for the carrier and the passenger based on the location and semantic tags that satisfy the meeting criteria. In addition, the computing system may generate instructions that may be used to facilitate navigation, including directions to the meeting location and a description of the meeting location. In this way, the disclosed technology may more efficiently determine a meeting location for use by individuals, including carriers and passengers.
[0028] Thus, the disclosed techniques may enable identification of locations where users can interact and / or meet more effectively and / or efficiently. Specifically, by determining the state of the physical environment, the behavior of users may be influenced to promote greater efficiency. Additionally, identifying suitable locations may improve safety by avoiding exposure of a party to situations where the environment is not suitable for them (e.g., where there may be locations that are not suitable for a party traveling on foot). The techniques may also allow users to more effectively and / or efficiently navigate to and identify a designated meeting location. In some examples, the meeting location may be referenced in a series of turn-by-turn navigation instructions that may be provided to a carrier in a vehicle and a passenger on foot. Thus, the described techniques may help a carrier more efficiently perform the technical task of driving a vehicle to a designated meeting location through a continuous and / or guided human-computer interaction process.
[0029] As another example, a computing system (e.g., a computing system in a vehicle) operated by a passenger who wishes to ride (e.g., ride by vehicle) to another location can access data associated with a geographic area (e.g., a combination of locally stored and / or remotely accessible data). The data may include location data (e.g., a map of an area including the location of the carrier and the passenger) and semantic labels indicating features of locations in the geographic area previously identified using image recognition (e.g., a machine learning model trained to detect features present in an image of the location). The computing system can then determine one or more meeting locations for the carrier and the passenger. For example, the computing system can determine a set of meeting locations that are safe and within a convenient walking distance for the passenger. The computing system can then provide the passenger with a set of instructions (e.g., audible instructions generated by the passenger's smart phone and / or text or graphical instructions provided on the display of the passenger's smart phone) to assist in navigation. At the same time, another set of instructions to the meeting location can be generated for use by the carrier. In this way, the disclosed technology provides various improvements in navigation and meeting coordination.
[0030] In some embodiments, the image content analysis system can analyze the semantic content of the image and provide a detailed description of the discovered entities in the form of semantic tags. The number of such semantic tags can be large (e.g., thousands), so the image content analysis system can have a very wide range in terms of what the image content analysis system can describe. In addition, these image content analysis systems can be configured to process a large number of imagery sets, for example, a street-level corpus including images captured from a street-level perspective. In addition, the system can utilize image content analysis of street-level images and / or other images (e.g., user-generated images and / or publicly accessible street-level images) for the purpose of determining meeting locations (including passenger boarding locations) that can be used for various transportation-related services and / or applications.
[0031] In one embodiment, semantic tags can be used to create semantic profiles of locations, which can be supplemented with location data (e.g., map data including information associated with known locations). These locations are not necessarily distinct points, but can be cells (e.g., areas of predetermined size) on a grid covering an area (e.g., a city, state, or the world) at a predetermined resolution (e.g., squares of two meters on each side). Semantic tags can be located and classified by such cells using the location of the image (e.g., a location associated with a street-level image), the semantic tag viewport, and the pose information of the image (e.g., the direction the camera was facing when the image was captured). Thus, items that are present in each cell and / or that may be unobstructed and visible from each cell can be determined.
[0032] To further reduce potential noise in the image content analysis response and take advantage of the density of street-level imagery (which in many areas may include multiple routes per street with many close images in a given route), the number of detections of a particular semantic tag may be compared to a threshold to determine the semantic tags of interest. In addition, the imaging time (e.g., the time the image was captured) of each semantic tag may be determined so that priority may be given to the closest detections. Additionally, for each detection, an image detection box may be used to determine the visibility and / or prominence of the detection from a given location.
[0033] In addition, semantic tags and / or units can be filtered so that only those semantic tags and / or units containing information useful for an application (e.g., an application for determining a meeting location for a carrier and a passenger) are stored. Specifically, a vocabulary of terms (e.g., keywords) specific to this application can be compiled, which includes words associated with a set of visual cues that are easily recognized by carriers and / or passengers. Using the disclosed technology, a large number of entities that can be used as visual cues can be identified. These entities can include visible business name signs, map features (e.g., parks and / or schools), landmarks (fountains, statues, and / or monuments), traffic signs and traffic elements ("stop" signs, traffic lights, and / or roundabouts), urban landscape elements (e.g., benches, lampposts, flower beds, and / or parking meters), road surface types (e.g., sidewalks), various colored stripe markings, parking areas (e.g., "loading areas", "no parking"), and / or architectural features (e.g., window types and / or canopies).
[0034] Given a query point and a radius, nearby cells with sufficient content can be collected and ranked based at least in part on proximity, ease of walking instructions (e.g., the number of turns required to travel to a location), the amount and visibility of semantic content that is easily recognizable to passengers on the ride compared to the amount and visibility of semantic content that is easily recognizable to the carrier, and / or a confidence score in the accuracy of the content detection (based in part on imaging time and in part on the number of identical semantic label detections).
[0035] In addition, the ease of recognition depends on the size and relative prominence of each semantic entity as it can be seen from the location itself and from other locations as the carrier and / or passengers approach a location along one or more recommended routes or paths. Therefore, for a given data set with a known range of values for each of these characteristics, a simple objective function can be constructed using specific weights for each feature. The cells with the highest scores and / or their adjacent neighboring cells (which can be used for directions context) can be selected, and their locations can be presented as options to the user (e.g., carriers and / or passengers). In addition, a concise description can be generated for each such location or retrieved from a database of selected text descriptions using, for example, natural language processing techniques.
[0036] In some embodiments, the disclosed technology may include a computing system (e.g., a geographic computing system) that may include one or more computing devices (e.g., a device having one or more computer processors and a memory that may store one or more instructions) that may send, receive, process, generate, and / or modify data (e.g., data including location data associated with a plurality of locations and semantic tags associated with an image) and / or one or more signals (e.g., electronic signals) that include one or more information patterns or structures that may be stored on one or more memory devices (e.g., one or more random access memory devices) and / or one or more storage devices (e.g., one or more hard disk drives and / or one or more solid-state memory drives). The data and / or one or more signals may be exchanged by the computing system with various other systems and / or devices including a plurality of service systems (e.g., one or more remote computing systems, one or more remote computing devices, and / or one or more software applications operating on one or more computing devices) that may send and / or receive data including location data and semantic tags associated with an image (e.g., a digital image associated with data including a geographic location, an image capture time, and / or one or more descriptions of one or more other features of the image). In some embodiments, the computing system (e.g., a geographic computing system) may include Figure 1 The device 102 and / or Figure 2 One or more features of the computing device 200 shown. In addition, the network computing system can be associated with one or more machine learning models, the one or more machine learning models including Figure 1 One or more features of one or more machine-learned models 120 are shown.
[0037] In addition, the computing system may include dedicated hardware (e.g., an application specific integrated circuit) and / or software that enables the computing system to perform one or more operations specific to the disclosed technology, including accessing location data (e.g., location data including the location of the carrier and the passenger) and semantic tags that may be associated with multiple images (e.g., accessing locally stored semantic tags and / or accessing the semantic tags by receiving the semantic tags from a remote computing device), determining a location suitable for use as a meeting location by the carrier and the passenger, and generating instructions associated with the meeting location (e.g., directions to the meeting location and / or a description of the meeting location).
[0038] The geographic computing system can access location data and / or multiple semantic tags. The location data may include information associated with multiple locations (e.g., geographic locations including latitude, longitude, and / or altitude of each location), and multiple locations include the locations of one or more parties (e.g., a set of geographic coordinates including latitude, longitude, and altitude) that may include a carrier (e.g., a driver of a vehicle) and the locations of passengers (e.g., passengers who need a vehicle to transport passengers to another location). In some embodiments, one or more parties may include one or more carriers and / or one or more passengers. In addition, each of the one or more carriers and / or one or more passengers may be located at various location combinations (e.g., several passengers at one location and several other passengers at another location).
[0039] In some embodiments, the location data may include information associated with multiple locations, including one or more maps that may indicate the locations of roads (e.g., streets, highways, and / or alleys), buildings, bodies of water, waterways, bridges, tunnels, overpasses, and / or underpasses. In addition, the location data may include information associated with traffic regulations (e.g., speed limits at a location, the location of traffic lights, and / or the location of stop signs), traffic flow patterns (e.g., the amount of vehicle or pedestrian traffic in an area), and / or the location of one or more hazards (including construction zones).
[0040] In some embodiments, the plurality of semantic tags may include information associated with one or more features of each of the plurality of locations that are visible from other locations of the plurality of locations, a geographic location associated with each of the plurality of locations, and / or a physical size (e.g., area in square meters) of each of the plurality of locations.
[0041] In addition, the plurality of semantic tags may include information associated with one or more features (e.g., one or more features of a location in the plurality of locations), including: visual features (e.g., the appearance of one or more objects at the location in the plurality of locations); physical dimensions of one or more objects at the location in the plurality of locations; and / or object identities (e.g., the identity of the object at each of the plurality of locations). In addition, the one or more features of the plurality of locations may include one or more features of the following items: objects at each of the plurality of locations, including buildings, utility poles, trees, mailboxes, benches, and / or trash cans; one or more surfaces at each of the plurality of locations (e.g., paved roads, lawns, and / or water); traffic markings (e.g., traffic signs and / or traffic signs on a road); and / or lighting features, including the location of street lights. In addition, each of the multiple semantic tags and / or the location data can be associated with information including: location (e.g., a street address and / or altitude, latitude, and longitude associated with the location and / or semantic tag); time of day (e.g., the time of day when the semantic tag was generated, modified, and / or last accessed); and / or date (e.g., the date when the semantic tag was generated, modified, and / or last accessed).
[0042] In some embodiments, each of the plurality of semantic tags may be associated with one or more features depicted by an image in the plurality of images. For example, each semantic tag may provide one or more semantic descriptions of one or more objects included in a scene depicted by one of the plurality of images. The plurality of images may include one or more digital images (e.g., two-dimensional images) of a portion of an environment (e.g., an image of a portion of a location in a plurality of locations). In addition, the plurality of images may be encoded in any type of image format, including a combination of raster images (e.g., a bitmap including an array of pixels) and / or vector images (e.g., a polygonal representation of an image based on coordinate positions of an x and y axis including a two-dimensional plane). The images may include still images, image frames from a movie, and / or other types of imagery (including LIDAR imagery and / or RADAR imagery).
[0043] Examples of digital image formats used by the plurality of images may include JPEG (Joint Photographic Experts Group), BMP (Bitmap), TIFF (Tagged Image File Format), PNG (Portable Network Graphics), and / or GIF (Graphics Interchange Format). Images may be collected from a variety of sources, including user-submitted images, images in the public domain (e.g., anonymous images obtained via web crawling), and / or street-level panoramic images.
[0044] For example, a geographic computing system may receive data including information associated with location data and / or multiple semantic tags via a communications network (e.g., a wireless and / or wired network including a LAN, WAN, or the Internet) over which one or more signals (e.g., electronic signals) and / or data may be sent and / or received.
[0045] The geographic computing system can identify location data and / or multiple semantic tags that satisfy one or more meeting criteria associated with the suitability of the location for the carrier and the passenger. For example, the geographic computing system can access data and / or information associated with the location data and / or multiple semantic tags, wherein the location data and / or multiple semantic tags include information associated with one or more characteristics of multiple locations that satisfy one or more meeting criteria associated with the suitability of the location for the carrier and the passenger. In addition, one or more characteristics of each of the multiple locations can be compared to each of the one or more meeting criteria that can be used to determine a meeting location suitable for the carrier and the passenger.
[0046] In addition, the suitability of a location for a carrier and / or passengers can be associated with one or more other characteristics of the location, including: the frequency of one or more features (e.g., how rarely the feature at the location occurs relative to other locations); the uniqueness of one or more features at the location (e.g., how unusual the feature at the location is); the prominence of one or more features at the location (e.g., the size and / or visibility of the feature at the location); the visual constancy of the features at the location (e.g., the extent to which the appearance of the features changes over time); the positional constancy of the features at the location (e.g., whether the feature will remain in the same place over time); and / or the context of the features at the location (e.g., one or more features that are in close proximity or contact with the features at the location).
[0047] Satisfaction of one or more meeting criteria may be based, for example, on a feature being prominent (e.g., a large neon sign mounted on a tall pole). Thus, in one example, an image content analysis system may be used to determine the prominence of each feature of a location, which may then be used to guide the selection of locations for use as meeting locations. As an example, for each location, an area surrounding the location (e.g., a ten meter by ten meter area) may be analyzed to identify which features associated with the location are most prominent (e.g., a size and / or height threshold may determine that a neon sign on a tall pole is prominent).
[0048] The geographic computing system may determine one or more meeting locations for the carrier and the passenger based at least in part on location data and / or multiple semantic tags that satisfy one or more meeting criteria. For example, the geographic computing system may determine that the one or more meeting locations include a predetermined number (e.g., one location) of multiple locations that satisfy the one or more meeting criteria. As another example, the geographic computing system may determine that the one or more meeting locations for the carrier and the passenger include a predetermined portion (e.g., all, half, or one quarter) of the multiple locations that satisfy the one or more meeting criteria. In addition, the geographic computing system may determine that the one or more meeting locations are within a predetermined distance (e.g., within ten meters) of the multiple locations that satisfy the one or more meeting criteria.
[0049] As an example, one or more meeting locations may be based at least in part on: the safety of the location (e.g., is the location safely accessible to pedestrians via crosswalks and / or sidewalks); the proximity of the location (e.g., how far the meeting location is from the carrier and / or passengers); the visibility of the location (e.g., how easily the carrier and / or passengers can see the location); and / or the accessibility of the location (e.g., how easily and / or conveniently pedestrians and / or vehicles can reach the meeting location).
[0050] The geographic computing system may generate one or more indications associated with at least one of the one or more meeting locations based at least in part on one or more characteristics of at least one of the one or more meeting locations. In addition, the amount of meeting locations included in at least one of the one or more meeting locations may be a predetermined amount (e.g., a single meeting location, a predetermined portion of one or more meeting locations, or a predetermined number of one or more meeting locations) or based at least in part on one or more criteria (e.g., a meeting location closest to the passenger).
[0051] As an example, the geographic computing system may generate one or more textual indications (including the address of the meeting location) and one or more visual indications (including images that may be displayed on display devices of the carrier and / or passengers).
[0052] In some embodiments, the one or more indications may include one or more text descriptions of the one or more meeting locations, one or more auditory descriptions of the one or more meeting locations, or one or more images of the one or more meeting locations (e.g., photographs of the one or more meeting locations).
[0053] In some embodiments, the geographic computing system may generate one or more descriptions of multiple locations within a predetermined distance of at least one location. For example, the geographic computing system may generate one or more descriptions of one or more features (e.g., features of a building) associated with multiple semantic tags, wherein the multiple semantic tags are associated with multiple locations within a predetermined distance of at least one location. As a further example, the geographic computing system may generate one or more natural language descriptions, including one or more natural language text descriptions (e.g., natural language text descriptions generated on a display device associated with the geographic computing system) and / or one or more audible descriptions (e.g., natural language audible descriptions using synthesized speech and / or pre-recorded speech).
[0054] In some embodiments, the geographic computing system may determine a confidence score for each of the plurality of semantic tags based at least in part on the number of times each of the plurality of semantic tags has been associated with a corresponding location in the plurality of locations. The confidence score for each of the plurality of semantic tags may be associated with the accuracy of each of the plurality of semantic tags (e.g., the degree to which the semantic tag correctly describes the state of the feature), respectively. For example, the geographic computing system may access data associated with the number of times a semantic tag associated with a feature is also associated with geographic data corresponding to the location (e.g., latitude and longitude information). The geographic computing system may then generate a confidence score corresponding to the number of times a semantic tag associated with a feature is also associated with geographic data corresponding to the location.
[0055] In addition, the geographic computing system can determine that the confidence score of each of the plurality of semantic tags satisfies one or more confidence score criteria. For example, the geographic computing system can determine that semantic tags having confidence scores below a confidence score threshold are not included in the determination of the one or more meeting locations.
[0056] In some embodiments, the geographic computing system identifying a plurality of tags that satisfy the one or more meeting criteria may include determining that a confidence score for each of the plurality of semantic tags satisfies the one or more confidence criteria.
[0057] In some embodiments, the confidence score of each semantic tag in the plurality of semantic tags may be negatively correlated with the age of the semantic tag. For example, the geographic computing system may access data associated with the time when each of the plurality of semantic tags was generated and determine that the confidence score of the semantic tag is reduced by an amount that increases proportionally with the age of the semantic tag.
[0058] In some embodiments, the geographic computing system may determine that each semantic tag in a plurality of semantic tags that corresponds to (e.g., matches or includes one or more similar characteristics) one or more keywords satisfies one or more meeting criteria. For example, the geographic computing system may determine that one or more meeting criteria are satisfied when a semantic tag is associated with one or more keywords including "restaurant," "bus stop," and / or "fountain." In addition, the correspondence of a semantic tag to a keyword may include an exact correspondence (e.g., a "fountain" semantic tag that matches the keyword "fountain") as well as a partial match (e.g., "Bellagio Fountains" may correspond to the keyword "fountain").
[0059] In some embodiments, the geographic computing system identifying a plurality of semantic tags that satisfy the one or more meeting criteria may include determining that the one or more meeting criteria are satisfied when each semantic tag in the plurality of semantic tags corresponding to the one or more keywords satisfies the one or more meeting criteria.
[0060] In some embodiments, one or more keywords may be associated with one or more features that are visually prominent (e.g., large and / or bright features) and / or one or more features that appear at a rate below a keyword rate threshold (e.g., rare features).
[0061] In some embodiments, the geographic computing system can determine the number of multiple semantic tags that correspond to one or more keywords. For example, the geographic computing system can compare the information in each of the multiple semantic tags with the one or more keywords to determine when the semantic tag corresponds to the keyword. In addition, each time the semantic tag corresponds to the keyword, the geographic computing system can increase the keyword counter by one. In this way, the geographic computing system can keep a count of the number of semantic tags that correspond to the one or more keywords.
[0062] In some embodiments, when the number of multiple semantic tags corresponding to one or more keywords exceeds a keyword threshold, the geographic computing system can determine that one or more meeting criteria are met. For example, when three or more semantic tags correspond to one or more keywords, the geographic computing system can determine that one or more meeting criteria are met.
[0063] In some embodiments, the geographic computing system determining the one or more meeting locations may include determining that the one or more meeting criteria are satisfied when a number of the plurality of semantic tags corresponding to the one or more keywords exceeds a keyword threshold.
[0064] In some embodiments, the geographic computing system can determine a corresponding visibility associated with each of the plurality of semantic tags. For example, the semantic tag can include information indicating that the location is located in an area with a large number of street lights that can improve the visibility of the location at night.
[0065] In addition, the geographic computing system can determine that the corresponding visibility associated with each of the plurality of semantic tags satisfies one or more visibility criteria. For example, the geographic computing system can determine that the visibility criteria are satisfied when the location includes at least one street light and / or is within a predetermined distance of a street light.
[0066] In some embodiments, the geographic computing system identifying a plurality of semantic tags that satisfy the one or more meeting criteria may include determining that a respective visibility associated with each of the plurality of semantic tags satisfies the one or more visibility criteria.
[0067] In some embodiments, determining that the visibility of one or more meeting locations satisfies one or more visibility criteria may include determining that the amount of light (e.g., the amount of light in lux) at the one or more meeting locations exceeds a light threshold, determining that the size (e.g., physical size) of one or more features associated with the one or more meeting locations exceeds a size threshold, determining that the number of locations from which a meeting location in the one or more meeting locations can be seen exceeds a location visibility threshold, or determining that the number of locations that are visible from a meeting location in the one or more meeting locations exceeds a vantage point visibility threshold.
[0068] In some embodiments, the geographic computing system may determine the location familiarity score for each of the multiple locations based at least in part on the number of times the carrier and / or passenger was previously within a predetermined distance of each of the multiple locations. In addition, the number of times the carrier and / or passenger was previously within a predetermined distance of each location may be determined based on information voluntarily provided by the carrier and / or passenger, respectively. For example, the carrier and / or passenger may receive an explicit request (e.g., "Can information about the location or area you have been to or passed through be collected for future use?") to request explicit permission from the carrier and / or passenger to use information about locations that the carrier and / or passenger has been to, passed through, and / or is within a predetermined distance of. In addition, any data and / or information associated with locations that the carrier and / or passenger has been to may be modified before storage or use to remove personally identifiable information. In addition, the stored information may be encrypted to prevent unauthorized access.
[0069] As an example, the location familiarity score may be increased each time the carrier and / or passenger has previously been within twenty meters of the location.
[0070] In addition, the geographic computing system can determine that the location familiarity scores of the one or more meeting locations satisfy one or more location familiarity criteria. For example, when the location familiarity score exceeds a location familiarity score threshold, the geographic computing system can determine that the one or more location familiarity criteria are satisfied.
[0071] In some embodiments, the geographic computing system determining the one or more meeting locations may include determining that the location familiarity scores of the one or more meeting locations satisfy one or more location familiarity criteria.
[0072] In some embodiments, the geographic computing system can determine that the location data and / or the plurality of semantic tags satisfy one or more safety criteria. For example, the geographic computing system can determine that the location data indicates that the one or more meeting locations are not in a construction zone, in the middle of a highway, or at the edge of a cliff. In addition, the geographic computing system can determine that the plurality of semantic tags do not include any semantic tags associated with potentially unsafe conditions, including falling rocks, wildlife, and / or heavy machinery.
[0073] In some embodiments, the geographic computing system can determine one or more paths from the location of the carrier and / or the passenger to one or more meeting locations. For example, the geographic computing system can access map information in the location data to determine one or more paths from the passenger's current location to a location via one or more streets.
[0074] Additionally, the geographic computing system may determine that the one or more meeting locations are associated with at least one of the one or more paths that satisfies one or more safety criteria (e.g., safety criteria that may be associated with the safety of passengers and / or carriers traveling to the one or more meeting locations). For example, the geographic computing system may determine that the one or more meeting locations are associated with (e.g., are connected to and reachable by) a path that has less than one road intersection per kilometer.
[0075] In some embodiments, the geographic computing system may determine at least one path from the location of the passenger and / or the carrier to at least one meeting location, the at least one path satisfying one or more path criteria associated with distance, complexity, and / or traffic. For example, the geographic computing system may determine at least one path from the location of the passenger to at least one meeting location, the at least one path satisfying a path criterion associated with a maximum distance of the path (e.g., at least one meeting location is closer to the passenger's departure location than the maximum distance). As another example, the geographic computing system may determine at least one path from the location of the passenger to at least one meeting location, the at least one path satisfying a path criterion associated with the complexity of the path (e.g., the number of turns in the path to at least one meeting location is less than a threshold value and does not deviate from the main road). As a further example, the geographic computing system may determine at least one path from the location of the passenger to at least one meeting location, the at least one path satisfying a traffic criterion associated with the amount of traffic along the path (e.g., a path along which the carrier and / or passenger will encounter less than a threshold amount of traffic to at least one meeting location).
[0076] In some embodiments, the geographic computing system may determine one or more meeting locations based at least in part on determining at least one path from a passenger and / or carrier location to at least one meeting location that satisfies one or more path criteria associated with distance, complexity, and / or safety of the path.
[0077] In some embodiments, the geographic computing system may determine the one or more meeting locations based at least in part on determining the one or more meeting locations associated with at least one of the one or more paths that satisfy the one or more safety criteria.
[0078] In some embodiments, satisfying one or more safety criteria may include: a ratio of road intersections on at least one path not exceeding a threshold ratio of road intersections, an amount of light along at least one path exceeding a threshold amount of light (e.g., the amount of light along the path may be determined based at least in part on the time of day and / or the amount of street lights along the path), a portion of at least one path includes a sidewalk, and / or at least one path does not pass through a hazardous object (e.g., a construction zone).
[0079] In some embodiments, the geographic computing system can determine one or more distances from the carrier's and / or passenger's location to the one or more meeting locations. For example, the geographic computing system can access map information in the location data to determine one or more distances (e.g., distances in kilometers) from the passenger's current location to each of the one or more meeting locations.
[0080] In addition, the geographic computing system can determine that the distance from the location of the passenger and / or the carrier to any of the one or more meeting locations is less than a distance threshold. For example, the geographic computing system can determine that the distance to any of the one or more meeting locations is less than an initial distance between the carrier's location and the passenger's location and / or a distance threshold based at least in part on the initial distance between the carrier and the passenger (e.g., the distance of the one or more meeting locations from the carrier or the passenger is no more than seventy-five percent of the initial distance between the carrier and the passenger).
[0081] In some embodiments, the geographic computing system may determine the one or more meeting locations based at least in part on determining that a distance from the passenger's location to any of the one or more meeting locations is less than a distance threshold.
[0082] In some embodiments, the geographic computing system may determine that one or more meeting criteria are met when the distance that the carrier and / or passenger can reach each of the one or more meeting locations is less than a reachable distance threshold. For example, the geographic computing system may access location data to determine the distance (e.g., in meters) around a location that is reachable by a carrier (e.g., a driver in a vehicle that is restricted to travel on a road) and a passenger (e.g., a pedestrian) (e.g., a distance within which a car can travel and a pedestrian can walk). In addition, the geographic computing system may determine that the distance of each of the one or more meeting locations is less than the reachable distance. For example, the geographic computing system may determine that the meeting location (e.g., a sidewalk and a bus stop that is accessible next to a road) is within a distance threshold of less than thirty meters.
[0083] In some embodiments, the geographic computing system may determine that the one or more meeting locations may include determining that a distance that the carrier and the passenger can reach each of the one or more meeting locations is less than a distance threshold.
[0084] In some embodiments, the geographic computing system may determine an ease of recognition score for each of the plurality of semantic tags, wherein the ease of recognition is associated with one or more features that are visually salient or one or more features that occur at a rate below a threshold rate of occurrence. For example, the geographic computing system may determine that a location with three visually salient features (e.g., three Christmas trees with holiday lights) will have a higher ease of recognition score than a location with one similarly decorated Christmas tree of the same size that is otherwise similar in appearance. As a further example, the geographic computing system may determine that a location with a visually salient feature (e.g., a large fountain with a horse sculpture) will have a higher ease of recognition score than a location that does not have a fountain and is otherwise similar in appearance.
[0085] In addition, the geographic computing system can determine that the recognition ease score of each of the plurality of semantic tags meets one or more recognition ease criteria. For example, the geographic computing system can determine that the recognition ease score of each of the plurality of semantic tags exceeds a recognition ease threshold.
[0086] In some embodiments, the geographic computing system identifying a plurality of semantic tags that satisfy the one or more meeting criteria may include determining that an ease of recognition score of the one or more meeting locations satisfies the one or more ease of recognition criteria.
[0087] In some embodiments, the geographic computing system may determine an arrangement of the one or more meeting locations (e.g., a manner of ranking and / or positioning the one or more meeting locations) based at least in part on the ease of recognition score associated with each of the one or more meeting locations. For example, the geographic computing system may determine an arrangement of the one or more meeting locations wherein one or more meeting locations associated with a higher ease of recognition score are positioned higher on a vertical axis of a display than one or more meeting locations having a lower ease of recognition score.
[0088] In addition, the geographic computing system can generate one or more indications including arranging the one or more meeting locations based at least in part on the arrangement of the one or more meeting locations. For example, the geographic computing system can generate a visual indication in which a higher meeting location associated with a higher ease of recognition score is displayed above a meeting location associated with a lower ease of recognition score.
[0089] In addition, the geographic computing system can receive input selecting at least one of the one or more meeting locations for the carrier or the passenger. For example, the geographic computing system can include a touch screen display that can display textual indications of the one or more meeting locations and receive touch input from the passenger to select one of the meeting locations as the meeting location where the passenger desires to meet the carrier.
[0090] The systems, methods, devices, apparatuses, and tangible, non-transitory computer-readable media of the disclosed technology can provide various technical effects and benefits, including improved meeting location determination for individuals (including carriers and / or passengers) for use in navigation and meeting coordination. Specifically, the disclosed technology can assist users (e.g., carriers in vehicles and / or passengers on foot) in performing technical tasks (e.g., driving a vehicle to a specified location and / or using a computing device as a navigation aid to walk to a specified location) with the aid of a continuous and / or guided human-computer interaction process. It can also provide benefits including improved resource utilization efficiency, security, and performance of communication networks.
[0091] The disclosed technology can improve the efficiency of resource consumption (e.g., fuel and electrical energy) by providing more effective instructions (e.g., directions to meeting locations that are more convenient to reach and closer to carriers (e.g., minimizing fuel consumption)) and passengers. For example, instructions that guide carriers and passengers to meeting locations that are convenient to reach can reduce missed meetings and waits by either party, thereby reducing the associated excess use of fuel or electrical energy caused by such missed meetings and waits.
[0092] Additionally, using safety criteria to determine meeting locations can improve passenger safety by directing passengers to areas that provide greater safety. For example, a meeting location that is in a well-lit area, accessible via a sidewalk, and not near hazards will improve passenger safety. Furthermore, a meeting location that is easily accessible to both the carrier and the passenger can reduce the incidence of potentially unsafe activities, including passengers walking into the roadway to meet the carrier.
[0093] In addition, the disclosed technology can improve the performance of communication network operations by more efficiently determining meeting locations for carriers and passengers. Meeting locations can be provided to more efficiently route traffic through the transportation network and avoid situations where communication network bandwidth is wasted due to ineffective navigation instructions (e.g., sending additional sets of instructions over the network when previous sets of instructions were not correctly followed). In this way, more efficiently providing navigation instructions can reduce the number of navigation instructions sent over the communication network and correspondingly reduce bandwidth utilization.
[0094] Thus, the disclosed technology can assist a driver of a vehicle in more effectively / efficiently performing the technical task of driving the vehicle to a specified location by means of a continuous and / or guided human-computer interaction process. In addition, the disclosed technology can provide a computing system that helps to more effectively identify meeting locations for navigation, geocoding, and transportation. The disclosed technology provides specific benefits of reduced network bandwidth utilization, more efficient use of fuel and energy, and increased safety, any of which can be used to improve the effectiveness of multiple services including navigation services and meeting coordination services.
[0095] Reference now Figure 1-Figure 14 , example embodiments of the present disclosure will be discussed in more detail. Figure 1 A diagram of an example system is depicted in accordance with an example embodiment of the present disclosure. System 100 includes computing device 102 , server computing system 130 , training computing system 150 , and one or more remote computing systems 170 communicatively connected and / or coupled via network 180 .
[0096] Computing device 102 may include any type of computing device, including, for example, a personal computing device (e.g., a laptop computing device or a desktop computing device), a mobile computing device (e.g., a smart phone or a tablet), a game console, a controller, a wearable computing device (e.g., a smart watch), an embedded computing device, and / or any other type of computing device.
[0097] The computing device 102 includes one or more processors 112 and memory 114. The one or more processors 112 can be any suitable processing device (e.g., a processor core, a microprocessor, an ASIC, an FPGA, a controller, a microcontroller, etc.), and can be one processor or multiple processors that are operably connected. The memory 114 can include one or more non-transitory computer-readable storage media, including RAM, ROM, EEPROM, EPROM, flash memory devices, magnetic disks, and combinations thereof. The memory 114 can store data 116 and instructions 118 that are executed by the processor 112 to cause the computing device 102 to perform operations.
[0098] In some embodiments, computing device 102 may perform one or more operations including accessing location data and / or multiple semantic tags. Location data may include information associated with multiple locations (including the location of the carrier and / or the location of the passenger). Multiple semantic tags may include information associated with one or more features of multiple locations. The operations performed by computing device 102 may also include identifying location data and multiple semantic tags that meet one or more meeting criteria associated with the suitability of the location for the carrier and the passenger. The operations performed by computing device 102 may include: determining one or more meeting locations for the carrier and the passenger based at least in part on location data and multiple semantic tags that meet one or more meeting criteria. In addition, the operations performed by computing device 102 may include: generating one or more indications associated with at least one of the one or more meeting locations based at least in part on one or more features of at least one of the one or more meeting locations.
[0099] In some implementations, the computing device 102 may store or include one or more machine learning models 120. For example, the one or more machine learning models 120 may include various machine learning models (including neural networks (e.g., deep neural networks)) or other types of machine learning models (including nonlinear models and / or linear models). The neural network may include a feedforward neural network, a recurrent neural network (e.g., a long short-term memory recurrent neural network), a convolutional neural network, or other forms of neural networks. Figure 1-Figure 14 Examples of one or more machine-learned models 120 are discussed.
[0100] In some embodiments, one or more machine-learned models 120 may be received from a server computing system 130 via a network 180, stored in a computing device memory 114, and then used or otherwise implemented by one or more processors 112. In some embodiments, the computing device 102 may implement multiple parallel instances of a single machine-learned model in the one or more machine-learned models 120 (e.g., performing parallel meeting location determination for carriers and passengers across multiple instances of the machine-learned model 120). More specifically, the one or more machine-learned models 120 may determine and / or identify a meeting location for use by a carrier and / or passenger based in part on various inputs including location data and / or semantic tags (e.g., semantic tags associated with one or more features of a location). In addition, the one or more machine-learned models 120 may determine one or more instructions to be provided in association with the identified meeting location.
[0101] Additionally or alternatively, one or more machine-learned models 140 may be included in or otherwise stored and implemented by a server computing system 130 that communicates with computing device 102 according to a client-server relationship. For example, machine-learned models 140 may be implemented by server computing system 140 as part of a network service (e.g., a suitable meeting location determination service). Thus, one or more machine-learned models 120 may be stored and implemented at computing device 102 and / or one or more machine-learned models 140 may be stored and implemented at server computing system 130.
[0102] The computing device 102 may also include one or more of the user input components 122 that may receive user input. For example, the user input component 122 may be a touch-sensitive component (e.g., a touch-sensitive display or touchpad) that is sensitive to the touch of a user input (e.g., a finger or a stylus). The touch-sensitive component may be used to implement a virtual keyboard. Other example user input components include a microphone, a traditional keyboard, or other means by which a user may provide user input.
[0103] The server computing system 130 includes one or more processors 132 and memory 134. The one or more processors 132 may include any suitable processing device (e.g., a processor core, a microprocessor, an ASIC, an FPGA, a controller, a microcontroller, etc.), and may include one processor or multiple processors that are operably connected. The memory 134 may include one or more non-transitory computer-readable storage media, including RAM, ROM, EEPROM, EPROM, flash memory devices, magnetic disks, etc., and combinations thereof. The memory 134 may store data 136 and instructions 138 that are executed by the processor 132 to cause the server computing system 130 to perform operations.
[0104] In some embodiments, the server computing system 130 may perform one or more operations including accessing location data and / or multiple semantic tags. The location data may include information associated with multiple locations (including the location of the carrier and / or the location of the passenger). Multiple semantic tags may include information associated with one or more features of multiple locations. The operations performed by the server computing system 130 may also include identifying location data and multiple semantic tags that meet one or more meeting criteria associated with the suitability of the location for the carrier and the passenger. The operations performed by the server computing system 130 may include: determining one or more meeting locations for the carrier and the passenger based at least in part on the location data and multiple semantic tags that meet the one or more meeting criteria. In addition, the operations performed by the server computing system 130 may include: generating one or more indications associated with at least one of the one or more meeting locations based at least in part on one or more features of at least one of the one or more meeting locations.
[0105] In addition, the server computing system 130 may perform image content analysis on one or more inputs (e.g., one or more images) provided to the server computing system 130. For example, the server computing system 130 may receive data including location data and / or associated semantic tags via the network 180. The server computing system 130 may then perform various operations (which may include using one or more machine learning models 140) to detect one or more features of the one or more images and / or associate semantic tags with the images. As a further example, the server computing system 130 may use object recognition techniques to detect one or more objects in the image (e.g., detecting pedestrians, vehicles, various buildings, fountains, street signs, and road markings) and generate semantic information based on the recognition of the objects. In another example, the server computing system 130 may receive data from one or more remote computing systems (e.g., one or more remote computing systems 170), which may include images that have been associated with semantic tags (e.g., images with semantic tags manually associated by a user and / or images with semantic tags generated by a user device). The data received by the server computing system 130 may then be stored (e.g., stored in an image repository) for later use by the computing system 130.
[0106] In some implementations, server computing system 130 includes or is otherwise implemented by one or more server computing devices. Where server computing system 130 includes multiple server computing devices, such server computing devices may operate according to a sequential computing architecture, a parallel computing architecture, or some combination thereof.
[0107] As described above, the server computing system 130 may store or otherwise include one or more machine learning models 140. For example, the one or more machine learning models 140 may include various machine learning models. Example machine learning models include neural networks or other multi-layer nonlinear models. Example neural networks include feedforward neural networks, deep neural networks, recurrent neural networks, and convolutional neural networks. Figure 1-Figure 14 Examples of one or more machine-learned models 140 are discussed.
[0108] Computing device 102 and / or server computing system 130 may train one or more machine-learned models 120 and / or 140 via interaction with a training computing system 150 that is communicatively connected and / or coupled via network 180. Training computing system 150 may be separate from server computing system 130, or may be part of server computing system 130.
[0109] The training computing system 150 includes one or more processors 152 and a memory 154. The one or more processors 152 can be any suitable processing device (e.g., a processor core, a microprocessor, an ASIC, an FPGA, a controller, a microcontroller, etc.), and can be one processor or multiple processors that are operably connected. The memory 154 can include one or more non-transitory computer-readable storage media, including RAM, ROM, EEPROM, EPROM, flash memory devices, disks, etc. and combinations thereof. The memory 154 can store data 156 and instructions 158 that are executed by the processor 152 to enable the training computing system 150 to perform operations. In some embodiments, the training computing system 150 includes or is otherwise implemented by one or more server computing devices.
[0110] The training computing system 150 may include a model trainer 160 that trains one or more machine-learned models 120 and / or one or more machine-learned models 140 stored at the computing device 102 and / or the server computing system 130, respectively, using various training or learning techniques, including, for example, backwards propagation of errors. In some embodiments, performing backwards propagation of errors may include performing truncated backpropagation over time. The model trainer 160 may perform a variety of generalization techniques (e.g., weight decay, dropout, etc.) to improve the generalization ability of the model being trained.
[0111] Specifically, the model trainer 160 can train one or more machine-learned models 120 and / or one or more machine-learned models 140 based on a set of training data 162. The training data 162 may include, for example, location data (e.g., data including information associated with pedestrian and vehicle interactions in a location), semantic data describing the location (e.g., semantic labels), and / or one or more features in an area. For example, the training data may include physical dimensions associated with the features, the proximity of the features to one or more reference points (e.g., other locations), the locations of the features (e.g., the latitude, longitude, and / or altitude of the features), and / or various metadata associated with the features (e.g., nicknames of the features, former names of the features, and / or alternate spellings of the features).
[0112] In some implementations, if the user has provided consent, the training examples may be provided by computing device 102. Thus, in such implementations, one or more machine-learned models 120 provided to computing device 102 may be trained by training computing system 150 on user-specific data received from computing device 102. In some cases, this process may be referred to as personalized models.
[0113] The model trainer 160 may include computer logic for providing the desired functionality. The model trainer 160 may be implemented in hardware, firmware, and / or software that controls a general purpose processor. For example, in some embodiments, the model trainer 160 includes a program file stored on a storage device, loaded into a memory, and executed by one or more processors. In other embodiments, the model trainer 160 includes one or more sets of computer executable instructions stored in a tangible computer readable storage medium including a RAM hard disk or an optical or magnetic medium.
[0114] In some embodiments, the training computing system 150 may perform one or more operations including accessing location data and / or multiple semantic tags. The location data may include information associated with multiple locations (including the location of the carrier and / or the location of the passenger). Multiple semantic tags may include information associated with one or more features of multiple locations. The operations performed by the training computing system 150 may also include identifying location data and multiple semantic tags that meet one or more meeting criteria associated with the suitability of the location for the carrier and the passenger. The operations performed by the training computing system 150 may include determining one or more meeting locations for the carrier and the passenger based at least in part on location data and multiple semantic tags that meet one or more meeting criteria. In addition, the operations performed by the training computing system 150 may include generating one or more indications associated with at least one of the one or more meeting locations based at least in part on one or more features of at least one of the one or more meeting locations.
[0115] Each of the one or more remote computing systems 170 includes one or more processors 172 and memory 174. The one or more processors 172 may include any suitable processing device (e.g., a processor core, a microprocessor, an ASIC, an FPGA, a controller, a microcontroller, etc.), and may include one processor or multiple processors that are operably connected. The memory 174 may include one or more non-transitory computer-readable storage media, including RAM, ROM, EEPROM, EPROM, flash memory devices, magnetic disks, etc., and combinations thereof. The memory 174 may store data 176 and instructions 178 that are executed by the processor 172 to cause the server computing system 170 to perform operations.
[0116] In some embodiments, one or more remote computing systems 170 include one or more computing devices or are otherwise implemented by one or more computing devices. Where one or more remote computing systems 170 include multiple computing devices, such computing devices may operate according to a sequential computing architecture, a parallel computing architecture, and / or some combination thereof. In addition, one or more remote computing systems 170 may be used to collect, generate, send, and / or receive one or more signals and / or data, the data including location data and / or one or more semantic tags that may be associated with one or more locations and / or one or more features of one or more locations. One or more remote computing systems 170 may include a smartphone device, which a user of the smartphone device may use to generate location data and / or semantic tags associated with one or more images captured by a camera of the smartphone device. For example, a user of a smartphone device may capture an image of a location (e.g., a restaurant) that includes the latitude and longitude of the location and geographic location information embedded in the image. The user may then manually annotate the image using a semantic tag including a location description (e.g., how crowded the location is at a particular time of day), which is then sent to a server computing system 170 including an image repository. The manually annotated images can be highly encrypted, anonymized (e.g., any personal information associated with the image is not collected or is deleted after receipt), and maintained in a privacy-enhancing manner for use by server computing system 170, which can provide the images and associated location data and / or semantic tags for use by the image content analysis portion of server computing system 170 and / or training computing system 150.
[0117] Network 180 may include any type of communications network, including a local area network (e.g., an intranet), a wide area network (e.g., the Internet), or some combination thereof, and may include any number of wired or wireless links. In general, communications through network 180 may be carried via any type of wired and / or wireless connection using a variety of communications protocols (e.g., TCP / IP, HTTP, SMTP, FTP), encodings or formats (e.g., HTML, XML), and / or protection schemes (e.g., VPN, secure HTTP, SSL).
[0118] Figure 1 An example computing system that can be used to implement the present disclosure is shown. Other computing systems may also be used. For example, in some embodiments, computing device 102 may include model trainer 160 and training data 162. In such embodiments, one or more machine-learned models 120 may be trained and used locally at computing device 102. In some such embodiments, computing device 102 may implement model trainer 160 to personalize one or more machine-learned models 120 based on user-specific data.
[0119] Figure 2 A diagram of an example device according to an example embodiment of the present disclosure is depicted. Computing device 200 may include one or more features of computing device 102, server computing system 130, and / or training computing system 150. In addition, computing device 200 may perform one or more actions and / or operations, including Figure 1 One or more actions and / or operations depicted in FIG. 1 are performed by computing device 102 , server computing system 130 , and / or training computing system 150 .
[0120] like Figure 2 As shown, the computing device 200 may include one or more memory devices 202, location data 204, semantic tag data 206, one or more interconnects 232, one or more processors 220, a network interface 222, one or more mass storage devices 224, one or more output devices 226, a sensor array 228, and one or more input devices 230.
[0121] One or more memory devices 202 may store information and / or data (e.g., location data 204 and / or semantic tag data 206), including information associated with processing of one or more instructions for performing one or more actions and / or operations, the one or more actions and / or operations comprising: accessing location data and / or semantic tags; identifying location data and / or semantic tags that satisfy meeting criteria; determining a meeting location for the carrier and the passenger based on the location data and / or semantic tags that satisfy the meeting criteria; and / or generating an indication based on characteristics of the meeting location.
[0122] Location data 204 may include Figure 1 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 111, 112, 113, 114, 115, 116, 117, 118, 119, 110, 111, 112, 113, 114, 115, 116
[0123] Semantic tag data 206 may include Figure 1 116, data 136, and / or data 156 depicted in FIG. 116. In addition, semantic tag data 206 may include information associated with one or more semantic tags (including semantic tags for describing one or more features of an image (e.g., an image retrieved from an image database)) and / or one or more maps (e.g., one or more maps in location data 204). In addition, semantic tag data may include multiple semantic tags, which may include Figure 8 One or more features of the multiple semantic tags described in the method 800 depicted in .
[0124] The one or more interconnects 210 may include one or more interconnects or buses that may be used to send and / or receive one or more signals (e.g., electronic signals) and / or data (e.g., location data 204 and / or semantic tag data 206) between components of the computing device 200, including one or more memory devices 202, one or more processors 220, a network interface 222, one or more mass storage devices 224, one or more output devices 226, a sensor array 228, and / or one or more input devices 230. The one or more interconnects 210 may be arranged or configured in different ways, including parallel connections or serial connections. In addition, the one or more interconnects 210 may include one or more internal buses that connect internal components of the computing device 200; and one or more external buses for connecting internal components of the computing device 200 to one or more external devices. For example, one or more interconnects 210 may include different interfaces, including Industry Standard Architecture (ISA), Extended ISA, Peripheral Components Interconnect (PCI), PCI Express, Serial AT Attachment (SATA), HyperTransport (HT), USB (Universal Serial Bus), Thunderbolt, IEEE 1394 (FireWire), and / or other interfaces for connecting components.
[0125] The one or more processors 220 may include one or more computer processors configured to run one or more instructions stored in the one or more memory devices 202. For example, the one or more processors 220 may include, for example, one or more general-purpose central processing units (CPUs), application specific integrated circuits (ASICs), and / or one or more graphics processing units (GPUs). In addition, the one or more processors 220 may perform one or more actions and / or operations, including one or more actions and / or operations associated with the location data 204 and / or the semantic tag data 206. For example, the one or more processors 220 may include a single-core or multi-core device, including a microprocessor, a microcontroller, an integrated circuit, and / or a logic device.
[0126] The network interface 222 can support network communications. For example, the network interface 222 can support communications via a network including a local area network and / or a wide area network (e.g., the Internet). One or more mass storage devices 224 (e.g., a hard drive and / or a solid state drive) can be used to store data including position data 204 and / or semantic tag data 206. One or more output devices 226 can include one or more display devices (e.g., an LCD display, an OLED display, and / or a CRT display), one or more light sources (e.g., LEDs), one or more speakers, and / or one or more tactile output devices.
[0127] The one or more input devices 230 may include one or more keyboards, one or more touch-sensitive devices (e.g., a touch screen display), one or more buttons (e.g., an on / off button, a yes / no button), one or more control levers (e.g., a joystick), one or more microphones, and / or one or more cameras (e.g., a camera that can be used to detect gestures and / or facial expressions).
[0128] The one or more memory devices 202 and the one or more mass storage devices 224 are shown separately, however the one or more memory devices 202 and the one or more mass storage devices 224 may be regions within the same memory module. The computing device 200 may include one or more additional processors, memory devices, network interfaces, which may be provided separately or on the same chip or board. The one or more memory devices 202 and the one or more mass storage devices 224 may include one or more computer-readable media, including but not limited to non-transitory computer-readable media, RAM, ROM, hard disk drives, flash drives, and / or other memory devices.
[0129] One or more memory devices 202 can store instruction sets for applications including operating systems, which can be associated with various software applications or data. One or more memory devices 202 can be used to operate various applications, including mobile operating systems developed specifically for mobile devices. In this way, one or more memory devices 202 can store instructions that allow software applications to access data including wireless network parameters (e.g., identification of wireless networks, quality of service), and call various services including phone calls, location determination (e.g., via global positioning services (global positioning service, GPS) or WLAN), and / or wireless network data call initiation services. In other embodiments, one or more memory devices 202 can be used to operate or run a general-purpose operating system that operates on mobile devices and fixed devices (e.g., smart phones, laptops, tablet computers and / or desktop computers).
[0130] Software applications that may be operated or run by computing device 200 may include Figure 1 Applications associated with the illustrated system 100. Additionally, software applications that may be operated or executed by the computing device 200 may include local applications and / or network-based applications.
[0131] In some embodiments, the computing device 200 may be associated with or include a positioning system (not shown). The positioning system may include one or more devices or circuits for determining the location of the computing device 200. For example, the positioning device may determine the actual or relative positioning by using a satellite navigation positioning system (e.g., GPS system, Galileo positioning system, Global Navigation Satellite System (GLONASS), Beidou Satellite Navigation and Positioning System), an inertial navigation system, a dead reckoning system, based on an IP address, by using triangulation and / or proximity to a cellular tower or Wi-Fi hotspot, beacons, etc., and / or other suitable techniques for determining positioning.
[0132] Figure 3 An example of determining a meeting location according to an example embodiment of the present disclosure is depicted. Figure 3 The outputs shown in may be generated and / or determined by a computing system or computing device including Figure 1 One or more features of computing device 102, server computing system 130, and / or training computing system 150 are depicted. Figure 3 As shown, image 300 includes architectural feature 302 , sculpture feature 304 , fountain feature 306 , and tree feature 308 .
[0133] Image 300 depicts a location having one or more features, including one or more features associated with various objects in the environment depicted by image 300 that have been identified by a computing system (e.g., a content analysis system including one or more machine-learned models trained to detect one or more features of input content that may include one or more images).
[0134] For example, the computing system may include one or more aspects of computing device 102, server computing system 130, and / or training computing system 150. In addition, one or more features of image 300 (including building feature 302, sculpture feature 304, fountain feature 306, and tree feature 308) may be associated with various semantic tags (e.g., including Figure 8The method 800 described in the embodiment of the present invention may be associated with a plurality of semantic tags of one or more features of the present invention, wherein various semantic tags may include descriptions of various aspects of one or more features. Figure 1 One or more features of the image 300 may be associated with one or more of the plurality of semantic tags, based on the image content analysis performed by the one or more machine learning models of one or more aspects of the one or more machine learning models 120 depicted. In addition, the image 300 may be associated with location data (e.g., metadata included in the image 300), which may include a location associated with the image 300 (e.g., latitude, longitude, and / or altitude), features depicted in the image 300 (e.g., fountain feature 306), and / or a vantage point from which the image 300 was captured.
[0135] In this example, architectural feature 302 is located in the background relative to other features of image 300 (e.g., sculpture feature 304 and fountain feature 306), and from the perspective of image 300, architectural feature 302 is partially obscured by objects including a tree represented by tree feature 308. In addition, architectural feature 302 may generate its own light (e.g., light external to architectural feature 302 and / or light emitted from interior portions of architectural feature 302 that are visible through windows of the building), which may increase the visibility of a location associated with architectural feature 302. The larger size of architectural feature 302 exceeds a size threshold (e.g., a height threshold and / or a footprint threshold), which may increase the prominence (e.g., visual prominence) of a location associated with architectural feature 302. The increase in prominence of architectural feature 302 may also result in an increase in the ease with which architectural feature 302 can be identified (e.g., the ease with which a computing system and / or a person (including a carrier or passenger) can identify the feature and / or a location associated with the feature). With respect to uniqueness, architectural feature 302 may be determined (e.g., based on a threshold value). Figure 1 300) are features of a frequently occurring type (e.g., there are many buildings throughout the city where image 300 is captured, including other similar-looking buildings present in image 300) and architectural feature 302 has lower uniqueness (e.g., various buildings throughout the area where image 300 is captured have similar size, shape, color, and appearance). Low uniqueness of architectural feature 302 may result in reduced ease of recognition of architectural feature 302.
[0136] Thus, based in part on visibility, prominence, uniqueness, and ease of identification, architectural feature 302 may be a type of feature that is less likely to be used as a meeting location than other locations that include more visible, prominent, unique, and / or more easily identified features.
[0137] Sculpture feature 304 is located in the foreground relative to other features of image 300 (e.g., architectural feature 302 and fountain feature 306), which may result in increased visibility of sculptural feature 304. Sculpture feature 304 is not illuminated by external lighting and cannot generate its own light, which may result in reduced visibility of sculptural feature 304 under certain conditions (e.g., at night and in fog, rain, snow, and / or hail), thereby resulting in reduced visibility of the location associated with sculptural feature 304. The size of sculptural feature 304 exceeds a size threshold (e.g., a height threshold and / or a volume threshold), which may result in increased prominence of sculptural feature 304. The prominence of sculptural feature 304 may result in increased ease of identification of sculptural feature 304. With respect to uniqueness, sculptural feature 304 may be determined to be a type of feature that does not occur frequently (e.g., there are no other similar-looking sculptures within a predetermined distance of sculptural feature 304). In addition, the unusual shape of sculptural feature 304 increases the uniqueness of sculptural feature 304. The increased uniqueness of sculptural feature 304 may result in greater ease of identification of sculptural feature 304 .
[0138] As such, based in part on visibility, visual prominence, uniqueness, and ease of identification, sculptural feature 304 may be a type of feature that is more likely to serve as a meeting location than other locations having less visible, prominent, unique, and / or less easily identified features.
[0139] Fountain feature 306 is not obscured by other objects depicted in image 300, which can result in increased visibility of fountain feature 306. Additionally, fountain feature 306 is illuminated by lighting equipment (not shown) within fountain feature 306 and can generate its own light, which can increase the visibility of fountain feature 306 in certain circumstances (e.g., at night and in fog, rain, snow, and / or hail), thereby increasing the visibility of the location depicted in image 300. The size of fountain feature 306 does not exceed a size threshold (e.g., a height threshold), which may reduce the prominence of the location depicted in image 300 due to the relatively lower height of fountain feature 306. However, fountain feature 306 has increased prominence due to the movement of its water nozzles, which can attract the attention of people (including carriers and / or passengers). The increase in prominence of fountain feature 306 can result in an increase in the ease of identification of fountain feature 306. With respect to uniqueness, fountain feature 306 can be determined (e.g., based on a threshold value). Figure 1 The increased uniqueness of fountain feature 306 may result in a greater ease of recognition of fountain feature 306.
[0140] As such, based in part on visibility, prominence, uniqueness, and ease of identification, fountain feature 306 may be a type of feature that is more likely to serve as a meeting location than other locations having less visible, prominent, unique, and / or less easily identified features.
[0141] Tree features 308 are not obscured by other objects, which may also increase the visibility of tree features 308. Additionally, tree features 308 are not illuminated by external lighting, which may reduce the visibility of tree features 308 in certain circumstances (e.g., at night and in fog, rain, snow, and / or hail), thereby not increasing the visibility of the location depicted in image 300. The size of tree features 308 exceeds a size threshold (e.g., a height threshold), which may make the location depicted in image 300 more prominent. However, the shape and color of tree features 308 may reduce the prominence of tree features 308. The reduced visual prominence of tree features 308 may result in a reduced ease of identification of tree features 308 (e.g., the ease with which a computing system and / or a person (including a carrier and / or a passenger) may identify the feature). With respect to uniqueness, tree features 308 may be determined (e.g., based on a threshold value). Figure 1 308) is a feature of a frequently occurring type (e.g., there are many other trees, including other similar-looking trees, in the area where image 300 was captured). The reduced uniqueness of tree feature 308 may result in a lower ease of recognition of tree feature 308.
[0142] Thus, based in part on visibility, prominence, uniqueness, and ease of identification, tree feature 308 may be a type of feature that is less likely to be used as a meeting location than other locations having more visible, prominent, unique, and / or more easily identified features.
[0143] Figure 4 An example of determining a meeting location according to an example embodiment of the present disclosure is depicted. Figure 4 The image shown in can be composed by including Figure 1 The computing system or computing device generates and / or determines one or more characteristics of the computing device 102, server computing system 130, and / or training computing system 150 depicted in FIG. Figure 4 As shown, image 400 includes crosswalk feature 402 , bicycle lane feature 404 , sidewalk feature 406 , sidewalk feature 408 , and street feature 410 .
[0144] Image 400 depicts a location having one or more features, including one or more features associated with various objects in the environment depicted in image 400 that have been identified by a computing system (e.g., a content analysis system including one or more machine-learned models trained to detect one or more features of input content that may include one or more images).
[0145] For example, the computing system may include one or more aspects of computing device 102, server computing system 130, and / or training computing system 150. In addition, one or more features of image 400 (including feature 402, bicycle lane feature 404, pedestrian walkway feature 406, pedestrian walkway feature 408, and street feature 410) may be associated with various semantic labels (e.g., including Figure 8 The various semantic tags may include descriptions of various aspects of one or more features. For example, based on one or more machine learning models (e.g., Figure 1 400 ), one or more features of the image 400 may be associated with one or more of the plurality of semantic tags. In addition, the image 400 may be associated with location data (e.g., metadata included in the image 400), which may include a location associated with the image 400 (e.g., latitude, longitude, and / or altitude), features depicted in the image 400 (e.g., crosswalk feature 402), and / or a vantage point from which the image 400 was captured.
[0146] In this example, crosswalk feature 402 represents a crosswalk between sidewalk feature 406 (located on one side of a street represented by street feature 410) and sidewalk feature 408 (located on the other side of the street represented by street feature 410, opposite sidewalk feature 406). Crosswalk feature 402 represents an area where it is safer for pedestrians to cross the street compared to an otherwise similar area lacking crosswalk feature 402. For example, crosswalk feature 402 may increase the safety of passengers walking and crossing the street represented by street feature 410.
[0147] Additionally, bike lane feature 404 represents a bike lane on the street represented by street feature 410. The presence of bike lane feature 404 may increase the safety of individuals (e.g., passengers riding bicycles to the meeting location) as they travel to the meeting location. Thus, the presence of bike lane feature 404 may increase the likelihood that the location associated with bike lane feature 404 will be used as the meeting location.
[0148] Figure 5An example of determining a meeting location according to an example embodiment of the present disclosure is depicted. Figure 5 The images shown in can be Figure 1 The computing system or computing device generates and / or determines one or more characteristics of the computing device 102, server computing system 130, and / or training computing system 150 depicted in FIG. Figure 5 As shown, image 500 includes parking meter feature 502 , parking space feature 504 , and sidewalk feature 506 .
[0149] Image 500 depicts a location having one or more features, including one or more features associated with various objects in the environment depicted in image 500 that have been identified by a computing system (e.g., a content analysis system including one or more machine-learned models trained to detect one or more features of input content that may include one or more images, as described herein).
[0150] For example, the computing system may include one or more aspects of computing device 102, server computing system 130, and / or training computing system 150. In addition, one or more features of image 500 (including parking meter feature 502, parking space feature 504, and sidewalk feature 506) may be associated with various semantic labels (e.g., including Figure 8 The various semantic tags may include descriptions of various aspects of one or more features. For example, based on one or more machine-learned models (e.g., Figure 1 500) performed using one or more machine learning models 120 depicted in the image content analysis, one or more features of the image 500 can be associated with one or more of the plurality of semantic tags. In addition, the image 500 can be associated with location data (e.g., metadata included in the image 500), which can include a location associated with the image 500 (e.g., latitude, longitude, and / or altitude), features depicted in the image 500 (e.g., parking meter feature 502), and / or a vantage point from which the image 500 was captured.
[0151] In this example, the location indicated by the sidewalk feature 506 represents a sidewalk that is very close (e.g., within a predetermined proximity distance threshold of ten meters) to the parking meter 502 and the parking space feature 504. The parking meter 502 and the parking space feature 504 can be identified as being associated with a location where a vehicle (e.g., a car operated by a carrier) can be parked. In addition, the sidewalk feature 506 can be identified as a location where an individual (e.g., a passenger) can safely wait (e.g., waiting for a carrier to pick up a passenger). The close proximity of the sidewalk feature 506, where the passenger can safely wait, to the parking meter feature 502 and the parking space feature 504, where the carrier can safely park, can indicate an increase in the accessibility (e.g., accessibility to the carrier and / or passenger) of the location depicted in the image 500. Therefore, the location depicted in the image 500 can be more likely to be determined as suitable for use as a meeting location for a carrier and a passenger.
[0152] Figure 6 An example of determining a meeting location according to an example embodiment of the present disclosure is depicted. Figure 6 The image shown can be composed by including Figure 1 The computing device or computing system generates and / or determines one or more characteristics of the computing device 102, server computing system 130, and / or training computing system 150 depicted in FIG. Figure 6 As shown, image 600 includes a carrier location 602 , a passenger location 604 , and a meeting location 606 .
[0153] Image 600 depicts a map of an area (the city of Denver) that includes a map associated with one or more semantic tags (e.g., Figure 8 A representation of a location associated with one or more semantic tags of a plurality of semantic tags described in the method 800 depicted in FIG. 8 , wherein the one or more semantic tags are associated with a particular feature (e.g., Figure 4 ). Features indicated by semantic labels (e.g., "bicycle lane" or "pedal lane") can indicate features within a predetermined distance of a location suitable for use as a meeting location (e.g., a meeting location for a carrier and a passenger). For example, a passenger can ride a bicycle to meet a carrier at a meeting location accessible by a bicycle lane. When determining locations suitable for use as meeting locations, bicycle lane features indicated by bicycle lane semantic labels can positively influence the identification of suitable locations. Therefore, all other things being equal, meeting location 606 is more likely to be identified as a meeting location than other areas that do not include bicycle lane features. For example, Figure 1 The computing device 102 depicted in FIG. 6 may select the meeting location 606 as the meeting location for the carrier located at the carrier location 602 and the passenger located at the passenger location 604 .
[0154] Additionally, the proximity of the meeting location 606 to the passenger location 604 may also increase the likelihood that the meeting location 606 is determined as the meeting location. For example, a passenger walking from the passenger location 604 may use a similar amount of time to travel to the meeting location 606 as a carrier driving (e.g., a car) to travel to the meeting location 606. Thus, depending on the mode of travel used by the carrier or the passenger, the distance to the meeting location may be weighted toward the party (e.g., the carrier or the passenger) that is closer and uses a slower mode of travel.
[0155] Figure 7 An example of determining a meeting location according to an example embodiment of the present disclosure is depicted. Figure 7 The output shown in can be replaced by including Figure 1 The computing system or computing device generates and / or determines one or more characteristics of the computing device 102, server computing system 130 and / or training computing system 150 depicted in the figure. As shown, Figure 7 Image 702 , image 704 , image 706 , sphere feature 708 , logo feature 710 , crosswalk feature 712 , and sign feature 714 are depicted.
[0156] Images 702 / 704 / 706 depict an image of a location having one or more features, including one or more features associated with various objects in the environment depicted in images 702 / 704 / 706 that have been identified by a computing system (e.g., a content analysis system including one or more machine learning models that are trained to detect one or more features of input content that may include one or more images). For example, the computing system may include one or more aspects of computing device 102, server computing system 130, and / or training computing system 150. In addition, one or more features of any of images 702 / 704 / 706 (including sphere feature 708, logo feature 710, crosswalk feature 712, and sign feature 714) may be associated with various semantic tags (e.g., including in Figure 8 The various semantic tags may include descriptions of various aspects of one or more features. For example, based on one or more machine-learned models (e.g., Figure 1One or more features of any one of the images 702 / 704 / 706 may be associated with one or more of the plurality of semantic tags, based on image content analysis performed using one or more machine learning models 120 depicted in . In addition, any one of the images 702 / 704 / 706 may be associated with location data (e.g., metadata included in the images 702 / 704 / 706), which may include a location (e.g., latitude, longitude, and / or altitude) associated with any one of the images 702 / 704 / 706, features depicted in the images 702 / 704 / 706 (e.g., spherical feature 708), and / or a vantage point from which any one of the images 702 / 704 / 708 was captured.
[0157] Sphere feature 708 is visible from a variety of different perspectives (including different perspectives captured in image 702 and image 706), which can increase the visibility of sphere feature 708 (e.g., visibility to carriers and / or passengers), thereby increasing the suitability of sphere feature 708 for use as a feature of a meeting location for carriers and passengers. In addition, sphere feature 708 is in close proximity to crosswalk feature 712 depicted in image 704, as indicated by sign feature 714 pointing in the direction of crosswalk feature 712. Being in close proximity to crosswalk feature 712 can increase safety associated with the meeting location (e.g., passengers can cross the road at the crosswalk rather than having to cross the road without a crosswalk). In this way, the visibility and positioning of sphere feature 708 can increase the likelihood that the associated location depicted in images 702 / 704 / 706 will be used as a meeting location.
[0158] Logo feature 710 is visible from a variety of different viewing angles (including different viewing angles captured in image 702 and image 706), which can increase the visibility of logo feature 710 (e.g., to carriers and / or passengers), thereby increasing the suitability of logo feature 710 for use in determining a meeting location for carriers and passengers. Additionally, logo feature 710 is in close proximity to crosswalk feature 712 depicted in image 704, as indicated by sign feature 714 pointing in the direction of crosswalk feature 712. The combination of sphere feature 708 and logo feature 710 can further increase the likelihood that a location containing those features will be determined to be a meeting location.
[0159] As an example, computing device 102 may generate instructions for carriers and / or passengers. The instructions generated by computing device 102 may include descriptions of sphere features 708 (e.g., "a half-meter high sphere"), logo features (e.g., "a circular logo with a smaller solid circle inside a larger hollow circle"), and crosswalk features (e.g., "a crosswalk next to the sphere").
[0160] Figure 8 A flow chart of an example method for determining a meeting location in an environment according to an example embodiment of the present disclosure is depicted. One or more portions of method 800 may be run or implemented on one or more computing devices or computing systems including, for example, computing device 102, server computing system 130, and / or training computing system 150. In addition, one or more portions of method 800 may be run or implemented as an algorithm on a hardware device or system disclosed herein. Figure 8 The steps performed in a specific order are depicted for illustration and discussion purposes. One of ordinary skill in the art using the disclosure provided herein will appreciate that the various steps of any method disclosed herein may be adjusted, modified, rearranged, omitted and / or expanded without departing from the scope of the disclosure.
[0161] At 802, method 800 may include accessing location data and / or multiple semantic tags. The location data may include information associated with multiple locations (e.g., geographic locations including latitude, longitude, and / or altitude of each location), multiple locations including locations (e.g., a set of geographic coordinates including latitude, longitude, and altitude) of one or more parties that may include a carrier (e.g., a driver of a vehicle) and locations of passengers (e.g., passengers who need a vehicle to transport passengers to another location). In some embodiments, one or more parties may include one or more carriers and / or one or more passengers. In addition, each of the one or more carriers and / or one or more passengers may be located at various location combinations (e.g., multiple passengers at one location and multiple other passengers at another location).
[0162] In some embodiments, the location data may include information associated with multiple locations, including one or more maps that may indicate the locations of roads (e.g., streets, highways, and / or alleys), buildings, bodies of water, waterways, bridges, tunnels, overpasses, and / or underpasses. In addition, the location data may include information associated with traffic regulations (e.g., speed limits at a location, locations of traffic lights, and / or locations of stop signs), traffic flow patterns (e.g., the volume of vehicle traffic or pedestrian traffic at a location and / or area), and / or the location of one or more hazards (including construction areas).
[0163] In some embodiments, the plurality of semantic tags may include information associated with one or more features of each of the plurality of locations that are visible from other locations of the plurality of locations, a geographic location associated with each of the plurality of locations, and / or a physical size (e.g., an area in square meters) of each of the plurality of locations.
[0164] In addition, the plurality of semantic tags may include information associated with one or more features (e.g., one or more features of a location in the plurality of locations), including: visual features (e.g., the appearance of one or more objects at the location in the plurality of locations); physical dimensions of one or more objects at the location in the plurality of locations; and / or object identities (e.g., the identity of the object at each of the plurality of locations). In addition, the one or more features of the plurality of locations may include one or more features of the following items: objects at each of the plurality of locations (including buildings, utility poles, trees, mailboxes, benches, and / or trash cans); one or more surfaces at each of the plurality of locations (e.g., paved roads, lawns, and / or water); traffic markings (e.g., traffic signs and / or traffic signs on a road); and / or lighting features (including the location of street lights). In addition, each of the multiple semantic tags and / or the location data can be associated with information including: location (e.g., a street address and / or altitude, latitude and / or altitude associated with the location and / or the semantic tag); time of day (e.g., the time of day when the semantic tag was created, modified and / or last accessed); and / or date (e.g., the date when the semantic tag was generated, modified and / or last accessed).
[0165] In some embodiments, each of the plurality of semantic tags may be associated with one or more features depicted by an image in the plurality of images. For example, each semantic tag may provide one or more semantic descriptions of one or more objects included in a scene depicted by one of the plurality of images. The plurality of images may include one or more digital images (e.g., two-dimensional images) of a portion of an environment (e.g., an image of a portion of a location in a plurality of locations). In addition, the plurality of images may be encoded in any type of image format including a combination of raster images (e.g., a bitmap including an array of pixels) and / or vector images (e.g., a polygonal representation of an image of a location based on coordinates of an x and y axis of a two-dimensional plane). The images may include still images, image frames from a movie, and / or other types of imagery, including LIDAR imagery and / or RADAR imagery.
[0166] Examples of digital image formats used by the plurality of images may include JPEG (Joint Photographic Experts Group), BMP (Bitmap), TIFF (Tagged Image File Format), PNG (Portable Network Graphic), and / or GIF (Graphics Interchange Format). Images may be collected from a variety of sources, including user-submitted images, images in the public domain (e.g., anonymous images obtained via web crawling), and / or street-level panoramic images.
[0167] As an example, computing device 102 may receive data including information associated with location data and / or multiple semantic tags via a communication network (e.g., a wireless network and / or a wired network including a LAN, WAN, or the Internet) over which one or more signals (e.g., electronic signals) and / or data may be sent and / or received.
[0168] At 804, method 800 may include identifying location data and / or multiple semantic tags that satisfy one or more meeting criteria associated with the suitability of the location for the carrier and the passenger. For example, computing device 102 may access data and / or information associated with location data and / or multiple semantic tags, the multiple semantic tags including information associated with one or more characteristics of multiple locations, the multiple locations satisfying one or more meeting criteria associated with the suitability of the location for the carrier and the passenger. In addition, computing device 102 may compare the location data and / or semantic data with data associated with the one or more meeting criteria to identify portions of the location data and / or multiple semantic tags that satisfy the one or more meeting criteria.
[0169] In some embodiments, the comparison of one or more features of each of the multiple locations (e.g., a comparison performed by computing device 102) may include a comparison of the location data and / or the content of each semantic tag to determine the extent to which the information in each semantic tag matches or is similar to the information in the data associated with the criteria. Satisfaction of one or more meeting criteria may include the content of the location data and / or semantic tags matching the meeting criteria (e.g., the semantic tag indicates the presence of a fountain), exceeding the meeting criteria by an amount (e.g., the height of the feature exceeds a predetermined height threshold measured in meters), being less than the meeting criteria by an amount (e.g., the distance to the passenger is less than a distance threshold measured in meters), and / or being within a range of the meeting criteria (e.g., the brightness is within a brightness range that is neither too bright nor too dim).
[0170] In addition, the suitability of a location for a carrier and / or passengers can be associated with one or more other characteristics of the location, including: the frequency of one or more features (e.g., how rarely the feature occurs at the location relative to other locations); the uniqueness of one or more features at the location (e.g., how unusual the feature is at the location); the prominence of one or more features at the location (e.g., the size and / or visibility of the feature at the location); the visual constancy of the features at the location (e.g., the extent to which the appearance of the features changes over time); the positional constancy of the features at the location (e.g., whether the feature remains in the same location over time); and / or the context of the features at the location (e.g., one or more features that are in close proximity or contact with the location features).
[0171] Meeting one or more meeting criteria may be based, for example, on a feature being prominent (e.g., a large neon sign mounted on a tall pole) or having sufficient capacity. Thus, in one example, an image content analysis system may be used to determine the capacity of a location, which may then be used to guide selection of a location for use as a meeting location. For example, for each location, the capacity of the location may be determined based at least in part on location data indicating physical dimensions of the location and / or semantic tags that may indicate an expected capacity at the location (e.g., a semantic tag indicating seating capacity of a recreation area).
[0172] At 806, method 800 may include determining one or more meeting locations for the carrier and the passenger based at least in part on the location data and / or the plurality of semantic tags that satisfy the one or more meeting criteria. For example, computing device 102 may determine one or more geographic locations associated with the location data and / or the plurality of semantic tags that satisfy the one or more meeting criteria. Computing device 102 may then determine the meeting location based on the one or more geographic locations associated with the plurality of semantic tags that satisfy the one or more meeting criteria using one or more techniques (e.g., clustering techniques).
[0173] As an example, one or more meeting locations may be determined based at least in part on a weighted combination of: the safety of the location (e.g., is the location safely accessible to pedestrians via crosswalks and / or sidewalks); the proximity of the location (e.g., how far the meeting location is from the carrier and / or passengers); the visibility of the location (e.g., how easily the location can be seen by the carrier and / or passengers); and / or the accessibility of the location (e.g., how easy and / or convenient it is for pedestrians and / or vehicles to reach the meeting location).
[0174] In some embodiments, one or more meeting locations may be based at least in part on the application of one or more secondary criteria to reduce the number of one or more meeting locations. For example, computing device 102 may determine that one or more meeting locations are within a predetermined distance (e.g., within fifteen meters) of the location of a crosswalk and / or a parking space for a vehicle. Therefore, any location more than 15 meters from a crosswalk or parking space cannot be determined as a meeting location. As another example, computing device 102 may determine that one or more meeting locations are within a predetermined distance (e.g., within one hundred meters) of the passenger's location. In this way, the distance that the passenger travels to the meeting location can be minimized.
[0175] At 808, method 800 may include generating one or more indications associated with at least one of the one or more meeting locations based at least in part on one or more characteristics of at least one of the one or more meeting locations. Additionally, the amount of meeting locations included in at least one of the one or more meeting locations may be a predetermined amount (e.g., a single meeting location, a predetermined portion of the one or more meeting locations, or a predetermined number of the one or more meeting locations) or based at least in part on one or more criteria (e.g., a meeting location closest to the passenger).
[0176] As an example, computing device 102 may generate one or more textual indications including an address of the meeting location and one or more visual indications including an image (e.g., a photograph of the meeting location retrieved from an image repository) that may be displayed on a display device of the carrier and / or the passenger.
[0177] In some embodiments, one or more instructions may include one or more text descriptions of one or more meeting locations (e.g., a natural language written description of the one or more meeting locations), one or more auditory descriptions of the one or more meeting locations (e.g., auditory descriptions generated by speakers of a device used by the carrier and / or the passenger), or one or more images of the one or more meeting locations (e.g., a photograph of the one or more meeting locations).
[0178] At 810, method 800 may include generating one or more descriptions of a plurality of locations within a predetermined distance of at least one location. For example, computing device 102 may generate one or more descriptions of one or more features (e.g., features of a building) associated with a plurality of semantic tags, the plurality of semantic tags being associated with the plurality of locations within a predetermined distance of the at least one location. As a further example, computing device 102 may generate one or more natural language descriptions, including one or more natural language text descriptions (e.g., natural language text descriptions generated on a display device associated with computing device 102) and / or one or more audible descriptions (e.g., natural language audible descriptions using synthesized speech and / or pre-recorded sounds).
[0179] Fig. 9 A flowchart of an example method for determining a meeting location in an environment according to an example embodiment of the present disclosure is depicted. One or more portions of method 900 may be run or implemented on one or more computing devices or computing systems including, for example, computing device 102, server computing system 130, and / or training computing system 150. One or more portions of method 900 may be run or implemented as an algorithm on a hardware device or system disclosed herein. In some embodiments, one or more portions of method 900 may be implemented as a Figure 8 In addition, one or more portions of method 900 may be performed as part of identifying location data and / or a plurality of semantic tags that satisfy one or more meeting criteria associated with the suitability of the location for the carrier and the passenger, as in Figure 8 As described in 804 of the method 800 depicted in . Fig. 9 The steps are depicted to be performed in a specific order for the purpose of illustration and discussion. One of ordinary skill in the art using the disclosure provided herein will understand that the various steps of any method disclosed herein may be adjusted, modified, rearranged, omitted and / or expanded without departing from the scope of the disclosure.
[0180] At 902, method 900 may include determining a confidence score for each of a plurality of semantic tags based at least in part on a number of times each of the plurality of semantic tags has been associated with a corresponding location in a plurality of locations. The confidence score for each of the plurality of semantic tags may be respectively associated with an accuracy of each of the plurality of semantic tags (e.g., the degree to which the semantic tag correctly describes a state of a feature). Additionally, a higher or increased confidence score may correspond to a semantic tag that more accurately describes an actual state of the feature.
[0181] For example, computing device 102 may access semantic tags associated with features (e.g., Figure 3The computing device 102 may then generate a confidence score corresponding to the number of times the semantic tag associated with the feature is also associated with the geographic data corresponding to the location (e.g., latitude and longitude information).
[0182] In some embodiments, the confidence score of each respective semantic tag in the plurality of semantic tags may be negatively correlated to the age of the semantic tag. For example, computing device 102 may access data associated with a time when each of the plurality of semantic tags was generated and determine that the confidence score of the semantic tag is increased by an amount proportional to how recently the semantic tag was generated (e.g., more recently generated semantic tags have higher confidence scores, indicating a higher confidence in the accuracy of the semantic tag).
[0183] At 904, method 900 may include determining when a confidence score of each of the plurality of semantic tags satisfies one or more confidence criteria. For example, computing device 102 may determine that only semantic tags with confidence scores exceeding a confidence score threshold are included in determining the one or more meeting locations.
[0184] Fig.10 A flowchart of an example method for determining a meeting location in an environment according to an example embodiment of the present disclosure is depicted. One or more portions of method 1000 may be run or implemented on one or more computing devices or computing systems including, for example, computing device 102, server computing system 130, and / or training computing system 150. In addition, one or more portions of method 1000 may be run or implemented as an algorithm on a hardware device or system disclosed herein. In some embodiments, one or more portions of method 1000 may be run or implemented as an algorithm on a hardware device or system disclosed herein. Figure 8 In addition, one or more parts of method 1000 may be implemented as described in Figure 8 The method 800 described in 804 of the method 800 depicted in , identifying location data and / or a plurality of semantic tags that satisfy one or more meeting criteria associated with the suitability of the location for the carrier and the passenger, is performed as a portion of. Fig.10 The steps are depicted to be performed in a specific order for the purpose of illustration and discussion. One of ordinary skill in the art using the disclosure provided herein will understand that the various steps of any method disclosed herein may be adjusted, modified, rearranged, omitted and / or expanded without departing from the scope of the disclosure.
[0185] At 1002, method 1000 may include determining when the location data and / or multiple semantic tags meet one or more safety criteria. For example, computing device 102 may use the location data to determine the location of each of the one or more meeting locations. In addition, computing device 102 may then determine that the location data indicates that the one or more meeting locations are not in a higher risk and / or less safe location (e.g., a construction area, the middle of a highway, or the edge of a cliff). As another example, computing device 102 may access multiple semantic tags to determine that the multiple semantic tags associated with the one or more meeting locations do not include any semantic tags associated with potentially unsafe conditions (including falling rocks, wild animals, and / or heavy machinery).
[0186] At 1004, method 1000 may include determining a respective visibility associated with each of the plurality of semantic tags. For example, computing device 102 may access a semantic tag (e.g., a semantic tag indicating the presence of a street light) that includes information indicating that the location is in an area with a large number of street lights, which may improve visibility of the location at night. As a further example, computing device 102 may access a semantic tag that includes information indicating that a feature (e.g., a fountain) is not obscured by other objects near the feature, which may improve visibility of the feature associated with the semantic tag.
[0187] In some embodiments, the visibility of each semantic tag can be associated with a numerical score or other numerical value (e.g., a visibility score or visibility value). In addition, a higher or increased visibility score or visibility value can correspond to a location that is more visible to carriers and / or passengers, and can correspond to a greater likelihood of using the location as a meeting location.
[0188] At 1006, method 1000 may include determining when a respective visibility associated with each of the plurality of semantic tags satisfies one or more visibility criteria. For example, computing device 102 may determine that a visibility criterion is satisfied when the location includes at least one street light and / or the illuminated building is within a predetermined distance of the street light.
[0189] In some embodiments, determining that the visibility of one or more meeting locations satisfies one or more visibility criteria may include determining that the amount of light (e.g., the amount of light in lux) at the one or more meeting locations exceeds a light threshold, determining that the size (e.g., physical size) of one or more features associated with the one or more meeting locations exceeds a size threshold, determining that the number of multiple locations from which the one or more meeting locations are visible exceeds a location visibility threshold, and / or determining that the number of the multiple locations that are visible from the one or more meeting locations exceeds a vantage point visibility threshold.
[0190] Fig.11 A flowchart of an example method for determining a meeting location in an environment according to an example embodiment of the present disclosure is depicted. One or more portions of method 1100 may be run or implemented on one or more computing devices or computing systems including, for example, computing device 102, server computing system 130, and / or training computing system 150. In addition, one or more portions of method 1100 may be run or implemented as an algorithm on a hardware device or system disclosed herein. In some embodiments, one or more portions of method 1100 may be implemented as a Figure 8 In addition, one or more portions of method 1100 may be implemented as described in Figure 8 The method 800 described in 804 of the method 800 depicted in , identifying a portion of the location data and / or multiple semantic tags that satisfy one or more meeting criteria associated with the suitability of the location for the carrier and the passenger. Fig.11 The steps are depicted to be performed in a specific order for the purpose of illustration and discussion. One of ordinary skill in the art using the disclosure provided herein will understand that the various steps of any method disclosed herein may be adjusted, modified, rearranged, omitted and / or expanded without departing from the scope of the disclosure.
[0191] At 1102, method 1100 may include determining that each of a plurality of semantic tags corresponding to one or more keywords (e.g., matches one or more similar characteristics, is synonymous with one or more similar characteristics, and / or includes one or more similar characteristics) satisfies one or more meeting criteria. For example, when a semantic tag matches one or more keywords including "restaurant," "bus stop," and / or "fountain," computing device 102 may determine that one or more meeting criteria are satisfied. In addition, the semantic tags corresponding to the keywords may include exact correspondences (e.g., a "fountain" semantic tag matches the keyword "fountain") as well as partial matches (e.g., "grand central" may correspond to the keyword "grand central station").
[0192] In some embodiments, one or more keywords may be associated with one or more features that are visually prominent (e.g., tall, large, and / or bright features) and / or one or more features that appear at a ratio below a keyword ratio threshold (e.g., unique, uncommon, or rare features).
[0193] At 1104, method 1100 may include determining a number of the plurality of semantic tags corresponding to the one or more keywords. For example, computing device 102 may compare information in each of the plurality of semantic tags to the one or more keywords to determine when the semantic tag corresponds to the keyword. Additionally, computing device 102 may increase a keyword counter by one each time the semantic tag corresponds to the keyword. In this manner, computing device 102 may maintain a count of the number of semantic tags corresponding to the one or more keywords.
[0194] At 1106, method 1100 may include determining that one or more meeting criteria are satisfied when the number of the plurality of semantic tags corresponding to the one or more keywords exceeds a keyword threshold. For example, computing device 102 may determine that one or more meeting criteria are satisfied when three or more of the semantic tags correspond to the one or more keywords.
[0195] In some embodiments, certain keywords may be weighted such that semantic tags corresponding to certain keywords may have a greater impact on satisfying the keyword threshold. For example, computing device 102 may determine that the word "fountain" may be weighted as three keywords alone and will satisfy a keyword threshold requiring three or more keywords.
[0196] Fig.12 A flowchart of an example method for determining a meeting location in an environment according to an example embodiment of the present disclosure is depicted. One or more portions of method 1200 may be run or implemented on one or more computing devices or computing systems including, for example, computing device 102, server computing system 130, and / or training computing system 150. One or more portions of method 1200 may be run or implemented as an algorithm on a hardware device or system disclosed herein. In some embodiments, one or more portions of method 1200 may be implemented as a Figure 8 In addition, one or more parts of method 1200 may be implemented as Figure 8 The method 800 depicted in 806 is performed by determining, at least in part, one or more meeting locations for the carrier and the passenger based on location data and / or a plurality of semantic tags that satisfy one or more meeting criteria. Fig.12 The steps are depicted to be performed in a specific order for the purpose of illustration and discussion. One of ordinary skill in the art using the disclosure provided herein will understand that the various steps of any method disclosed herein may be adjusted, modified, rearranged, omitted and / or expanded without departing from the scope of the disclosure.
[0197] At 1202, method 1200 may include determining a location familiarity score for each of the plurality of locations based at least in part on the number of times the carrier or passenger was previously within a predetermined distance of each of the plurality of locations. In addition, the determination of the location familiarity score, the locations that the carrier and / or passenger has been to, and / or the number of times the carrier and / or passenger was previously within a predetermined distance of each location may be based in part on information voluntarily provided by the carrier and / or passenger, respectively. For example, the carrier and / or passenger may receive an explicit request (e.g., "Can information about the locations or areas you have been to or passed through be recorded for future use?") to request explicit permission from the carrier and / or passenger to use information about locations that the carrier and / or passenger has been to, passed through, and / or is within a predetermined distance of. In addition, any data and / or information associated with locations that the carrier and / or passenger has been to may be modified before storage or use to remove personally identifiable information.
[0198] In some embodiments, a higher or increased location familiarity score may correspond to a location that is more familiar to the carrier and / or passenger, and may correspond to a greater likelihood of using the location as a meeting location. For example, in some embodiments, a user may provide computing device 102 with authorization to access a list of previously visited locations. For example, computing device 102 may access location history data associated with locations that the carrier and / or passenger has previously visited. In addition, computing device 102 may compare locations that the carrier and / or passenger has previously visited to multiple locations, and determine that the location familiarity score will increase each time the carrier and / or passenger has previously been within twenty-five meters of a location.
[0199] At 1204, method 1200 may include determining when the location familiarity score of the one or more meeting locations satisfies one or more location familiarity criteria. For example, a higher location familiarity score may correspond to a higher familiarity of the location on the carrier and / or passenger side. Additionally, computing device 102 may determine that one or more location familiarity criteria are satisfied when the location familiarity score exceeds a location familiarity score threshold.
[0200] Fig.13 A flowchart of an example method for determining a meeting location in an environment according to an example embodiment of the present disclosure is depicted. One or more portions of method 1300 may be run or implemented on one or more computing devices or computing systems including, for example, computing device 102, server computing system 130, and / or training computing system 150. One or more portions of method 1300 may be run or implemented as an algorithm on a hardware device or system disclosed herein. In some embodiments, one or more portions of method 1300 may be implemented as a Figure 8In addition, one or more parts of method 1300 may be implemented as Figure 8 The method 800 depicted in 806 is performed by determining, at least in part, one or more meeting locations for the carrier and the passenger based on location data and / or a plurality of semantic tags that satisfy one or more meeting criteria. Fig.13 The steps are depicted to be performed in a specific order for the purpose of illustration and discussion. One of ordinary skill in the art using the disclosure provided herein will understand that the various steps of any method disclosed herein may be adjusted, modified, rearranged, omitted and / or expanded without departing from the scope of the disclosure.
[0201] At 1302, method 1300 may include determining one or more paths from the location of the carrier and / or passenger to the one or more meeting locations. For example, computing device 102 may access map information in the location data to determine one or more paths from the current location of the carrier and / or passenger to the one or more meeting locations via one or more streets. In addition, the one or more paths may be based on roads, streets, and / or sidewalks between the location of the carrier and / or passenger and the one or more meeting locations that do not violate traffic regulations.
[0202] At 1304, method 1300 may include determining when the one or more meeting locations are associated with at least one of the one or more paths that satisfy the one or more safety criteria. For example, computing device 102 may determine that the one or more meeting locations are associated with (e.g., connected to and reached by) a path that does not traverse any hazards (e.g., a construction area).
[0203] In some embodiments, satisfying one or more safety criteria may include a ratio of road intersections on at least one path not exceeding a threshold ratio of road intersections, an amount of light along at least one path exceeding a threshold amount of light (e.g., the amount of light along the path may be determined at least in part based on the time of day and / or the amount of street lights along the path), a portion of at least one path including a sidewalk, and / or at least one path not passing through a hazard (e.g., a construction zone).
[0204] At 1306, method 1300 may include determining one or more distances from the location of the carrier and / or the passenger to the one or more meeting locations. For example, computing device 102 may access map information in the location data to determine one or more distances (e.g., one or more distances in meters) from the passenger's current location to each of the one or more meeting locations.
[0205] At 1308, method 1300 may include determining when a distance from the location of the passenger and / or the carrier to any of the one or more meeting locations is less than a distance threshold. For example, computing device 102 may determine that the distance to any of the one or more meeting locations is less than an initial distance between the location of the carrier and the location of the passenger and / or a distance threshold based at least in part on the initial distance between the carrier and the passenger (e.g., the distance of the one or more meeting locations from the carrier or the passenger is no more than fifty percent of the initial distance between the carrier and the passenger). As another example, computing device 102 may determine that the one or more meeting locations are no more than two hundred meters from the location of the passenger.
[0206] At 1310, method 1300 may include determining when one or more meeting criteria are met when the distance that the carrier and / or passenger can reach each of the one or more meeting locations is less than a reachable distance threshold. For example, computing device 102 may access location data to determine a distance (e.g., in meters) around a location that is reachable by both the carrier (e.g., a driver in a vehicle limited to traveling on a road) and the passenger (e.g., a pedestrian) (e.g., a distance within which a car can legally drive and a pedestrian can walk without encountering vehicle traffic). In addition, computing device 102 may determine that the distance to each of the one or more meeting locations is less than the reachable distance. For example, computing device 102 may determine that the meeting location (e.g., a reachable bus stop next to a sidewalk and road) is within a distance threshold of less than twenty meters.
[0207] Fig.14 A flowchart of an example method for determining a meeting location in an environment according to an example embodiment of the present disclosure is depicted. One or more portions of method 1400 may be executed or implemented on one or more computing devices or computing systems including, for example, computing device 102, server computing system 130, and / or training computing system 150. One or more portions of method 1400 may be executed or implemented as an algorithm on a hardware device or system disclosed herein. In some embodiments, one or more portions of method 1400 may be implemented as a Figure 8 In addition, one or more parts of method 1400 may be implemented as Figure 8 The method 800 described in 804 of the method 800 depicted in , identifying location data and / or a plurality of semantic tags that satisfy one or more meeting criteria associated with the suitability of the location for the carrier and the passenger, is performed as a portion of. Fig.14The steps are depicted to be performed in a specific order for the purpose of illustration and discussion. One of ordinary skill in the art using the disclosure provided herein will understand that the various steps of any method disclosed herein may be adjusted, modified, rearranged, omitted and / or expanded without departing from the scope of the disclosure.
[0208] At 1402 , method 1400 may include determining an ease of recognition score for each of a plurality of semantic tags, wherein the ease of recognition is associated with one or more features that are visually salient or one or more features that appear at a rate below an occurrence rate threshold.
[0209] In some embodiments, a higher or increased ease of recognition score may correspond to a location that is more easily recognized by the carrier and / or passengers, and may correspond to a greater likelihood of using the location as a meeting location. For example, computing device 102 may determine that a location with three visually salient features (e.g., three large bronze statues) will have a higher ease of recognition score than an otherwise visually similar location with one bronze statue of the same size. As a further example, computing device 102 may determine that a location with a visually salient feature (e.g., a large fountain with lion sculptures) will have a higher ease of recognition score than an otherwise visually similar location that lacks a fountain.
[0210] At 1404, method 1400 may include determining when an ease of recognition score for each of the plurality of semantic tags satisfies one or more ease of recognition criteria. For example, computing device 102 may compare each ease of recognition score to an ease of recognition score threshold and determine that each semantic tag that exceeds the ease of recognition threshold satisfies the one or more ease of recognition criteria.
[0211] At 1406, method 1400 may include determining an arrangement of the one or more meeting locations (e.g., a manner of ranking and / or positioning the one or more meeting locations) based at least in part on the ease of recognition score associated with each of the one or more meeting locations. For example, computing device 102 may determine an arrangement of the one or more meeting locations in which one or more meeting locations associated with a higher ease of recognition score are positioned higher on a vertical axis of the display than one or more meeting locations with a lower ease of recognition score.
[0212] As another example, computing device 102 may determine an arrangement of one or more meeting locations wherein audible indications of one or more meeting locations associated with higher ease of recognition scores are generated via a speaker device before one or more meeting locations having lower ease of recognition scores.
[0213] At 1408, method 1400 may include generating one or more indications, including arranging the one or more meeting locations based at least in part on the placement of the one or more meeting locations. For example, computing device 102 may generate an audible indication, wherein higher meeting locations associated with higher ease scores are generated before meeting locations associated with lower ease scores.
[0214] At 1410, method 1400 may include receiving input to select at least one of the one or more meeting locations for a carrier or a passenger. For example, computing device 102 may include a touch screen display that may display textual indications of the one or more meeting locations and may receive touch input from a passenger to select one of the meeting locations as the meeting location at which the passenger desires to meet with the carrier. As another example, computing device 102 may include a microphone that may detect speech from the passenger indicating that the passenger selects one of the one or more meeting locations as the meeting location at which the passenger desires to meet with the carrier.
[0215] The technology discussed herein relates to servers, databases, software applications, and other computer-based systems, as well as actions taken and information sent to and from such systems. The inherent flexibility of computer-based systems allows for a variety of possible configurations / combinations and divisions of tasks and functions between and among components. For example, the processes discussed herein may be implemented using a single device or component or multiple devices or components working in combination. Databases and applications may be implemented on a single system or distributed across multiple systems. Distributed components may operate sequentially or in parallel.
[0216] Although the subject matter has been described in detail for various specific example embodiments of the subject matter, each example is provided by way of explanation rather than limitation of the present disclosure. Those skilled in the art, after understanding the foregoing, can easily make changes, modifications and equivalents to these embodiments. Therefore, the present disclosure does not exclude such modifications, modifications and / or additions to the subject matter that are obvious to those of ordinary skill in the art. For example, a feature shown or described as part of one embodiment can be used together with another embodiment to produce yet another embodiment. Therefore, it is intended that the present disclosure covers such changes, modifications and equivalents.
Claims
1. A computer-implemented method for providing navigation instructions, the method comprising: accessing, by a computing system comprising one or more processors, location data and a plurality of semantic tags associated with one or more features of a plurality of locations, wherein the location data is associated with the plurality of locations and includes a location of a carrier and a location of a passenger, and wherein the plurality of semantic tags are determined using image content analysis to identify one or more features in a plurality of images associated with the plurality of locations; identifying, by the computing system, location data and a plurality of semantic tags that satisfy one or more meeting criteria associated with suitability of the location for the carrier and the passenger, wherein identifying the plurality of semantic tags comprises: determining, by the computing system, a confidence score for each of the plurality of semantic labels based at least in part on a number of times each semantic label in the plurality of semantic labels has been associated with a corresponding location in the plurality of locations, wherein the confidence score for each of the plurality of semantic labels is respectively associated with an accuracy of each of the plurality of semantic labels; and including, by the computing system, semantic tags from the plurality of tags having corresponding confidence scores that satisfy one or more confidence criteria as semantic tags identified as satisfying one or more meeting criteria; determining, by the computing system, one or more meeting locations for the carrier and the passenger based at least in part on the plurality of semantic tags and the location data identifying the plurality of semantic tags that satisfy the one or more meeting criteria and have corresponding confidence scores that satisfy one or more confidence criteria; and One or more indications associated with at least one of the one or more meeting locations are generated by the computing system based at least in part on one or more characteristics of the at least one meeting location.
2. The computer-implemented method of claim 1 , wherein: The confidence score of each corresponding semantic tag is negatively correlated with the age of the semantic tag.
3. A computer-implemented method according to any one of the preceding claims, wherein: Identifying, by the computing system, the location data and the plurality of semantic tags that satisfy one or more meeting criteria associated with suitability of a location for the carrier and the passenger includes: determining, by the computing system, a respective visibility associated with each semantic tag in the plurality of semantic tags; and A determination is made by the computing system when a respective visibility associated with each semantic tag in the plurality of semantic tags satisfies one or more visibility criteria.
4. The computer-implemented method of claim 3, wherein: Determining, by the computing system, when respective visibilities of the one or more meeting locations satisfy one or more visibility criteria includes: determining, by the computing system, when an amount of light at the one or more meeting locations exceeds a light threshold, determining, by the computing system, when a size of one or more features associated with the one or more meeting locations exceeds a size threshold, determining, by the computing system, when a number of a plurality of locations from which a meeting location of the one or more meeting locations is visible exceeds a location visibility threshold, or A determination is made by the computing system when a number of a plurality of locations visible from a meeting location of the one or more meeting locations exceeds a vantage point visibility threshold.
5. The computer-implemented method of claim 1 or 2, wherein: Identifying, by the computing system, the location data and the plurality of semantic tags that satisfy one or more meeting criteria associated with suitability of a location for the carrier and the passenger includes: It is determined by the computing system that each semantic tag corresponding to the one or more keywords satisfies the one or more meeting criteria.
6. The computer-implemented method of claim 5, wherein: The one or more keywords are associated with one or more features that are visually prominent or one or more features that appear at a ratio below a keyword ratio threshold.
7. The computer-implemented method of claim 1 or 2, wherein: Identifying, by the computing system, the location data and the plurality of semantic tags that satisfy one or more meeting criteria associated with suitability of a location for the carrier and the passenger includes: Determining, by the computing system, a number of a plurality of semantic tags corresponding to the one or more keywords; and When the number of the plurality of semantic tags corresponding to the one or more keywords exceeds a keyword threshold, the computing system determines that the one or more meeting criteria are satisfied.
8. The computer-implemented method of claim 1 or 2, wherein: Determining, by the computing system, one or more meeting locations for the carrier and the passenger based at least in part on the location data and the plurality of semantic tags that satisfy the one or more meeting criteria includes: determining, by the computing system, one or more routes from the passenger's location to the one or more meeting locations based at least in part on the location data; and A determination is made by the computing system when the one or more meeting locations are associated with at least one of the one or more paths that satisfies one or more safety criteria.
9. The computer-implemented method of claim 8, wherein: Determining, by the computing system, when the one or more meeting locations are associated with at least one of the one or more paths that satisfies one or more safety criteria includes: determining, by the computing system, when a ratio of road intersections on the at least one path does not exceed a threshold ratio of road intersections, determining, by the computing system, when an amount of light along the at least one path exceeds a threshold amount of light, determining, by the computing system, when a portion of the at least one path includes a sidewalk, or A determination is made by the computing system when the at least one path does not traverse a hazard.
10. The computer-implemented method of claim 1 or 2, wherein: Determining, by the computing system, one or more meeting locations for the carrier and the passenger based at least in part on the location data and the plurality of semantic tags that satisfy the one or more meeting criteria includes: determining, by the computing system, a plurality of distances from the passenger's location to the one or more meeting locations; and A determination is made by the computing system when a distance from the passenger's location to any of the one or more meeting locations is less than a distance threshold.
11. The computer-implemented method of claim 1 or 2, wherein: Determining, by the computing system, one or more meeting locations for the carrier and the passenger based at least in part on the location data and the plurality of semantic tags that satisfy the one or more meeting criteria includes: The computing system determines when the one or more meeting criteria are met when a distance within which the carrier and the passenger can reach each of the one or more meeting locations is less than an accessible distance threshold.
12. The computer-implemented method of claim 1 or 2, wherein: Determining, by the computing system, one or more meeting locations for the carrier and the passenger based at least in part on the location data and the plurality of semantic tags that satisfy the one or more meeting criteria includes: determining, by the computing system, a location familiarity score for each of the plurality of locations based at least in part on a number of times the carrier or the passenger has previously been within a predetermined distance of each of the plurality of locations; and A determination is made by the computing system when the location familiarity score of the one or more meeting locations exceeds a location familiarity score threshold.
13. The computer-implemented method of claim 1 or 2, wherein: The one or more indications include one or more textual descriptions of the one or more meeting locations, one or more auditory descriptions of the one or more meeting locations, or one or more images of the one or more meeting locations.
14. One or more tangible, non-transitory computer-readable media storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising: accessing location data and a plurality of semantic tags associated with one or more features of a plurality of locations, wherein the location data is associated with the plurality of locations and includes a location of a carrier and a location of a passenger, and wherein the plurality of semantic tags are determined using image content analysis to identify one or more features in a plurality of images associated with the plurality of locations; identifying location data and a plurality of semantic tags that satisfy one or more meeting criteria associated with suitability of the location for the carrier and the passenger, wherein identifying the plurality of semantic tags comprises: determining, by a computing system, a confidence score for each of the plurality of semantic labels based at least in part on a number of times each semantic label in the plurality of semantic labels has been associated with a corresponding location in the plurality of locations, wherein the confidence score for each of the plurality of semantic labels is respectively associated with an accuracy of each of the plurality of semantic labels; and including, by the computing system, semantic tags from the plurality of tags having corresponding confidence scores that satisfy one or more confidence criteria as semantic tags identified as satisfying one or more meeting criteria; determining one or more meeting locations for the carrier and the passenger based at least in part on the plurality of semantic tags and the location data identifying the plurality of semantic tags that satisfy the one or more meeting criteria and have corresponding confidence scores that satisfy one or more confidence criteria; and One or more indications associated with at least one of the one or more meeting locations are generated based at least in part on the one or more characteristics of the at least one meeting location.
15. One or more tangible, non-transitory computer-readable media according to claim 14, wherein: Identifying the location data and the plurality of semantic tags that satisfy one or more meeting criteria associated with the suitability of a location for the carrier and the passenger includes: determining a respective ease of recognition score for each of the plurality of semantic tags, wherein the respective ease of recognition score is associated with one or more features that are visually salient or one or more features that appear at a rate below a threshold rate of appearance; and A determination is made as to when a corresponding ease of recognition score for each semantic tag in the plurality of semantic tags satisfies one or more ease of recognition criteria.
16. The one or more tangible, non-transitory computer-readable media of claim 15, further comprising: determining a placement of the one or more meeting locations based at least in part on a respective ease of recognition score associated with each of the plurality of semantic tags; generating one or more indications based at least in part on the arrangement of the one or more meeting locations; as well as Input is received to select at least one of the one or more meeting locations for the carrier or the passenger.
17. A computing system comprising: one or more processors; One or more non-transitory computer-readable media storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising: accessing location data and a plurality of semantic tags associated with one or more features of a plurality of locations, wherein the location data is associated with the plurality of locations and includes a location of a carrier and a location of a passenger, and wherein the plurality of semantic tags are determined using image content analysis to identify one or more features in a plurality of images associated with the plurality of locations; identifying location data and a plurality of semantic tags that satisfy one or more meeting criteria associated with suitability of the location for the carrier and the passenger, wherein identifying the plurality of semantic tags comprises: determining, by the computing system, a confidence score for each of the plurality of semantic labels based at least in part on a number of times each semantic label in the plurality of semantic labels has been associated with a corresponding location in the plurality of locations, wherein the confidence score for each of the plurality of semantic labels is respectively associated with an accuracy of each of the plurality of semantic labels; and including, by the computing system, semantic tags from the plurality of tags having corresponding confidence scores that satisfy one or more confidence criteria as semantic tags identified as satisfying one or more meeting criteria; determining one or more meeting locations for the carrier and the passenger based at least in part on the plurality of semantic tags and the location data identifying the plurality of semantic tags that satisfy the one or more meeting criteria and have corresponding confidence scores that satisfy one or more confidence criteria; and One or more indications associated with at least one of the one or more meeting locations are generated based at least in part on the one or more characteristics of the at least one meeting location.
18. The computing system of claim 17, wherein: Generating one or more indications associated with at least one of the one or more meeting locations based at least in part on one or more characteristics of the at least one meeting location includes: One or more descriptions of a plurality of locations within a predetermined distance of the at least one meeting location are generated.
19. The computing system according to claim 17 or 18, wherein: Identifying the location data and the plurality of semantic tags that satisfy one or more meeting criteria associated with the suitability of a location for the carrier and the passenger includes: A determination is made when the location data and the plurality of semantic tags satisfy one or more security criteria.
20. One or more tangible, non-transitory computer-readable media storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the method of any one of claims 1 to 13.
21. A computing system comprising: one or more processors; One or more non-transitory computer-readable media storing instructions that, when executed by the one or more processors, cause the one or more processors to perform the method according to any one of claims 1 to 13.
Citation Information
Patent Citations
Method of operating a navigation system to provide a pedestrian route
EP2372305A2
Method and system for identifying meeting points
US20180023967A1
Dynamic geolocation optimization of pickup paths using curb segment data
US20180328747A1
Method and apparatus to enable location-based meeting
US9264849B1
