WO2019028853A1

WO2019028853A1 - Method, crowdsourcing platform and system for building three-dimensional image model of object

Info

Publication number: WO2019028853A1
Application number: PCT/CN2017/097153
Authority: WO
Inventors: 骆磊
Original assignee: 深圳前海达闼云端智能科技有限公司
Priority date: 2017-08-11
Filing date: 2017-08-11
Publication date: 2019-02-14
Also published as: CN108064397A; CN108064397B

Abstract

Disclosed is a method, a crowdsourcing platform and a system for building a three-dimensional image model of an object. The method comprises: acquiring a model building task published by an initiating end, wherein the model building task comprises relevant information about a target object, and the model building task is used to request to build a 3D model of the target object; analyzing, according to the relevant information, photographs to be taken which are required for building the 3D model of the target object and photographing coordinate information of the photographs to be taken; generating at least one sub-task according to the photographs to be taken and the photographing coordinate information, wherein the at least one sub-task comprises a task requesting to photograph the target object; assigning the at least one sub-task to at least one destination end; and receiving photographs uploaded by the at least one destination end which are used to respond to the at least one sub-task, wherein the photographs are used to build the 3D model of the object.

Description

Method for building object three-dimensional image model, crowdsourcing platform and system

Technical field

The present disclosure relates to the field of object modeling, and in particular, to a method, a crowdsourcing platform and a system for establishing a three-dimensional image model of an object.

Background technique

At present, the demand for 3D modeling of objects in games, creative design and other fields is increasing, which promotes the development of 3D modeling technology. However, the modeling of existing 3D modeling techniques for larger objects (such as large buildings) still requires a professional team to operate, which is costly and inefficient.

Crowdsourcing refers to the practice of a company or organization outsourcing a work task performed by employees in the past to a non-specific (and usually large) mass network in a free and voluntary manner. How to solve the technical problems existing in the above 3D modeling technology by means of crowdsourcing is not considered by those skilled in the art.

Summary of the invention

The main objective of the embodiments of the present disclosure is to provide a method, a crowdsourcing platform and a system for establishing a three-dimensional image model of an object, which are used to solve the technical problem of high cost and low efficiency of the existing 3D modeling technology.

In order to achieve the above object, a first aspect of an embodiment of the present disclosure provides a method for establishing a three-dimensional image model of an object, the method being applied to a crowdsourcing platform, the method comprising:

Obtaining a modeling task issued by the initiator, where the modeling task includes related information of the target object, and the modeling task is used to request to establish a 3D model of the target object;

And analyzing, according to the related information, a photo to be photographed required to establish a 3D model of the target object, and shooting coordinate information of the photo to be photographed;

Generating at least one subtask according to the photograph to be taken and the photographing coordinate information, the at least one subtask comprising requesting a photographing task on the target object;

Assigning the at least one subtask to at least one destination;

Receiving a photo uploaded by the at least one destination for responding to the at least one subtask, the photo being used to establish a 3D model of the object.

A second aspect of an embodiment of the present disclosure provides a crowdsourcing platform, including:

a task acquisition module, configured to acquire a modeling task issued by the initiator, where the modeling task includes related information of the target object, where the modeling task is used to request to establish a 3D model of the target object;

a modeling analysis module, configured to analyze, according to the related information, a photo to be photographed required to establish a 3D model of the target object, and shooting coordinate information of the photo to be photographed;

a subtask generating module, configured to generate at least one subtask according to the photograph to be photographed and the photographing coordinate information, where the at least one subtask includes a task of requesting photographing the target object;

a task dispatching module, configured to assign the at least one subtask to at least one destination;

And a receiving module, configured to receive a photo uploaded by the at least one destination for responding to the at least one subtask, where the photo is used to establish a 3D model of the object.

A third aspect of an embodiment of the present disclosure provides a non-transitory computer readable storage medium including one or more programs for performing a first The method described in the aspects.

A fourth aspect of the embodiments of the present disclosure provides a crowdsourcing platform, comprising: the non-transitory computer readable storage medium of the third aspect;

One or more processors for executing a program in the non-transitory computer readable storage medium.

A fifth aspect of the embodiments of the present disclosure provides an object modeling system, comprising: the crowdsourcing platform of the second aspect or the fourth aspect, and an initiator end and a destination end, wherein the initiator end and the destination end The destination end is a terminal device registered with the crowdsourcing platform.

The above technical solutions can achieve at least the following technical effects:

After obtaining the modeling task issued by the initiator, the crowdsourcing platform can automatically analyze the photos needed to create the 3D model of the target object and the photos to be taken according to the related information of the target object described in the modeling task. Coordinates, and, based on the analysis results, the crowdsourcing platform can divide the modeling task into a plurality of subtasks and assign them to multiple destinations for execution. In this way, for the initiator, it only needs to publish the modeling task describing the relevant information of the target object on the crowdsourcing platform, and does not need to undertake the photo shooting work required to establish the model, thereby reducing the cost and, in mass, When the destination is willing to accept and execute the task assigned by the crowdsourcing platform, the efficiency of taking the photos required for modeling will be greatly improved, thereby improving the efficiency of object modeling.

DRAWINGS

In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings used in the description of the embodiments will be briefly described below. It is obvious that the drawings in the following description are the disclosure of the present disclosure. For some embodiments, other drawings may be obtained from those of ordinary skill in the art without departing from the drawings.

FIG. 1 is a schematic flowchart of a method for establishing an object model in a crowdsourcing manner according to an embodiment of the present disclosure;

2 is a schematic diagram of a scene in which a target end captures a target object according to an embodiment of the present disclosure;

FIG. 3 is a schematic diagram of an implementation environment according to an embodiment of the present disclosure;

4 is a schematic flow chart of a method for establishing an object model in a crowdsourcing manner according to the implementation environment shown in FIG. 3;

FIG. 5 is a schematic structural diagram of a crowdsourcing platform according to an embodiment of the present disclosure;

FIG. 6 is a schematic structural diagram of another crowdsourcing platform according to an embodiment of the present disclosure;

FIG. 7 is a schematic structural diagram of still another crowdsourcing platform according to an embodiment of the present disclosure.

Detailed ways

The specific embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. It is to be understood that the specific embodiments described herein are not to be construed

An embodiment of the present disclosure provides a method for establishing a three-dimensional image model of an object, where the method is applied to a crowdsourcing platform, where the crowdsourcing platform may be a network server or a network server cluster provided by an operator, as shown in FIG. 1 . Methods include:

S101. Acquire a modeling task issued by the initiator, where the modeling task includes related information of the target object, and the modeling task is used to request to establish a 3D model of the target object.

The initiator may be a user terminal device that has been registered on the crowdsourcing platform, such as a user's personal computer. The target object may be an object that can be uniquely specified on the map by the user, such as a large building, or an iconic plant. The related information of the target object may include at least one of the following: a picture of the target object, a name of the target object, a geographic coordinate of the target object, a height information of the target object, and an area information of the target object.

For example, the user of the initiator can provide the information of the target object to be modeled as much as possible when publishing the modeling task, especially for large buildings, where the user can describe the geographical coordinates of the building in detail in the modeling task. Ground height and area. In the specific implementation, the crowdsourcing platform can provide a variety of fillable items for users to fill in. For some necessary information, the crowdsourcing platform can also set the corresponding fillable items as mandatory. In addition, for some landmark buildings, such as the Eiffel Tower, the user can also just fill in the name of the landmark or upload a picture of the landmark, and the crowdsourcing platform can search the logo for other landmarks based on the name or picture. Related Information.

In a possible implementation manner of the embodiment of the present disclosure, after receiving the modeling task issued by the initiator, the crowdsourcing platform may further determine, according to the related information of the target object, whether the target object specified by the initiator is unique, if the crowdsourcing The platform can uniquely determine the target object according to the description of the target object in the modeling task, and then perform the subsequent step S102; if the crowdsourcing platform cannot determine an object on the map according to the description of the target object in the modeling task, or If you identify multiple objects on the map, you can The feedback message asks the user to provide more information of the target object or requires the user to select the target object among the plurality of objects currently determined by the crowdsourcing platform until the crowdsourcing platform uniquely determines the target object, and performs the subsequent step S102.

S102. Analyze, according to the related information of the target object, a photo to be taken required to establish a 3D model of the target object, and shooting coordinate information of the photo to be taken.

The related information of the target object may include information about the target object described in the modeling task issued by the originating end, and may further include other related information of the acquired target object by the crowdsourcing platform based on the description in the modeling task.

It is worth noting that the 3D model of the object needs to use the photos of the various orientations and angles of the object. Based on the contour of the object in the 3-dimensional space, the number of photos needed to build the model of the object and the shooting position of each photo can be analyzed. The 3D coordinate range. Therefore, in the case that the related information of the target object includes the geographical coordinates, the height information, and the footprint information of the target object, the photograph to be taken and the photographing coordinate information of the photograph to be photographed are analyzed and analyzed to establish a 3D model of the target object. Specifically, the location of the target object on the 3D map is determined according to the geographic coordinates, the 3D map includes contour information of the target object, and the 3D model of the completed target object is determined according to the contour information and the height information and the footprint information analysis. a 3D coordinate range of a photographing position of a photographing position to be photographed and a photographing position of each photograph to be photographed, which may be part or all of contour information of the target object, the photographing coordinate information including the 3D coordinate range.

S103. Generate at least one subtask according to the to-be-captured photo and the photographing coordinate information, where the at least one subtask includes requesting a photographing task on the target object.

That is to say, each subtask requires taking a picture of the target object at the position specified by the shooting coordinate information to obtain a photo to be taken. The real scene of the target object is filled into the outline of the photo to be photographed required to establish the 3D model of the target object.

In the specific implementation, the request for taking a photo to be taken can be taken as a sub-rule. The number of subtasks is also N, and in another possible implementation, multiple photos to be taken with the shooting coordinates within a preset distance may also be photographed. Required as a subtask.

S104. Assign the at least one subtask to at least one destination.

The destination end may be another user terminal device that has been registered on the crowdsourcing platform, such as a user's smartphone, tablet, or smart camera.

For example, the dispatching of the subtask by the crowdsourcing platform may be specifically allocated according to the current positioning of the destination end. In order to facilitate the destination end user to quickly reach the specified coordinate position to complete the task, the crowdsourcing platform may obtain the positioning of the destination end, if the purpose If the positioning position of the end is smaller than the predetermined distance from the shooting position of a subtask, the subtask is assigned to the destination.

S105. Receive a photo uploaded by the at least one destination for responding to the at least one subtask, where the photo is used to establish a 3D model of the object.

Using the above method, after obtaining the modeling task issued by the originating platform, the crowdsourcing platform can automatically analyze the photo taken to establish the 3D model of the target object according to the related information of the target object described in the modeling task, and The coordinates of the photo to be taken, and the crowdsourcing platform can divide the modeling task into a plurality of subtasks and assign them to multiple destinations based on the analysis results. In this way, for the initiator, it only needs to publish the modeling task describing the relevant information of the target object on the crowdsourcing platform, and does not need to undertake the photo shooting work required to establish the model, thereby reducing the cost and, in mass, When the destination is willing to accept and execute the task assigned by the crowdsourcing platform, the efficiency of taking the photos required for modeling will be greatly improved, thereby improving the efficiency of object modeling.

Specifically, the crowdsourcing platform assigns the at least one subtask to the at least one destination end by the following two methods:

The first sub-task is the location information of the first sub-task, and the location information of the first sub-task meets the predetermined distance condition, and the first sub-task is any one of the at least one sub-task. Subtask, if the first subtask's shooting coordinate information and the first item The location information of the terminal meets the predetermined distance condition, and the first sub-task is assigned to the first destination.

That is to say, the crowdsourcing platform can obtain the positioning information of each destination end in real time, and determine whether each destination end is located within a predetermined range of a certain shooting coordinate, if the first destination end is currently located at the shooting coordinates of the first subtask Within the predetermined range, the first subtask is assigned to the first destination.

After receiving the subtask, the destination end can mark the shooting coordinates of the subtask on the map of the destination end, which is convenient for the user to arrive. If the destination end is in the case of not completing the subtask, it is far away from the predetermined range of the subtask shooting coordinates. The crowdsourcing platform can cancel the subtask assigned to the destination.

The second method is to obtain the positioning information of the second destination end and the shooting device parameters of the second destination end, and determine, according to the shooting coordinate information of the second subtask and the shooting device parameter, whether the shooting device of the second destination end meets the preset shooting requirement, The second subtask is any one of the at least one subtasks, if the photographing device of the second destination meets the preset photographing requirement, and the photographing coordinate information of the second subtask and the second target end If the positioning information satisfies the predetermined distance condition, the second subtask is assigned to the second destination.

That is to say, in order to ensure that the destination end can capture the photo of the 3D model for establishing the target object, the crowdsourcing platform can request the shooting device of the destination to meet certain conditions, such as taking a photo of the photographing device, when assigning the subtask. The resolution is high enough. Therefore, when the destination end registers the crowdsourcing platform, the relevant parameters of the destination camera can be reserved, so that the crowdsourcing platform can filter the destination of the subtask by shooting device parameters when dispatching the task.

For example, the photographing device parameter includes an angle of view α of the camera and a maximum photographing resolution m*n, the photographing coordinate information of the second subtask is (x1, y1), and the coordinates of the center point of the target object is (x2, Y2), the determining whether the photographing device of the second destination meets the preset photographing requirement according to the photographing coordinate information of the second subtask and the photographing device parameter may include:

Calculating the lateral pixel value x and the vertical pixel value y of the target object by the photographing device of the second destination end according to the following formula:

x=m*c/(2*tan(α/2)*√((x1–x2) ² +(y1–y2) ² ));

y=n*b/(2*tan(α/2)*√((x1–x2) ² +(y1–y2) ² ));

Where c is the width of the target object relative to the shooting coordinate direction of the second subtask, b is the height of the target object relative to the shooting coordinate direction of the second subtask, and determines whether the horizontal pixel value x is greater than a preset horizontal direction. a threshold and determining whether the longitudinal pixel value y is greater than a preset longitudinal threshold. If the horizontal pixel value x is greater than the preset horizontal threshold and the vertical pixel value y is greater than the preset vertical threshold, it is determined that the photographing device of the second destination meets the preset shooting requirement.

For example, as shown in FIG. 2, the coordinate position specified by the subtask is O(x1, y1), the coordinate of the center point E of the target object facing the coordinate position O is (x2, y2), and the target object faces the coordinate position O. The width value of the direction is c, and the maximum angle of view of the camera (the angle of view α of the camera) and the intersection of the target object in the width direction are A and B. Thus, from the triangle principle, the distance between OE points is:

D1=√((x1–x2) ² +(y1–y2) ² )

The distance between the AB points is:

D2=2*d1*tan(α/2).

In turn, the horizontal pixel values of the target object in the photo are:

x=m*(c/d2)=m*c/(2*tan(α/2)*√((x1–x2) ² +(y1–y2) ² )).

Similarly, the vertical pixel values of the target object in the photo can be obtained:

y=n*b/(2*tan(α/2)*√((x1–x2) ² +(y1–y2) ² )).

It is to be noted that the coordinates of the point closest to the coordinate position O of the target object may be replaced by the coordinates of the center point, which is not limited in the embodiment of the present disclosure.

In another possible implementation manner of the embodiment of the present disclosure, the crowdsourcing platform may also detect the photograph taken by the destination end online, and determine whether the photograph taken by the destination end satisfies the modeling requirement. In this case, the parameters of the photographing device can be reserved without the destination end.

The subtasks generated by the crowdsourcing platform are described below. Optionally, the child assigned to the destination The information may include photographing guide information for guiding the destination end to reach a position specified by the photographing coordinate information, and guiding the destination end to display an outline of the photograph to be taken of the subtask.

For example, after receiving the subtask, the destination end may display the coordinate position of the subtask requesting the photograph on the map, and may also display the navigation between the current position of the user and the coordinate position of the subtask requesting the photograph, so that the user can arrive. Specify the coordinate position. In the case where the subtask includes the outline of the photo to be taken (the outline is the outline of the target object in the photo to be photographed), the destination end can also display the outline on the photographing interface of the photographing device, so that the user Simply take a picture of the target object into the outline and take a picture.

Optionally, for the destination end having the remote control function, the crowdsourcing platform may also send a remote control request to the destination end, and after receiving the response sent by the destination end for confirming the permission of the remote control, the crowdsourcing platform remote control purpose The end completes the subtask received by the destination. For example, if the destination end is a drone, the shooting of one or more coordinate positions can be directly performed under the takeover operation of the crowdsourcing platform, without the intervention of the destination user.

Further, after obtaining the photo uploaded by the destination end, the crowdsourcing platform may determine whether the received photo for responding to the third subtask meets a preset condition, where the third subtask is in the at least one subtask Any subtask, if the photo used to respond to the third subtask meets the preset condition, marking the third subtask as completed.

It is worth noting that the crowdsourcing platform can assign the same subtask to multiple destinations when dispatching subtasks, so that when the subtask is completed on a certain destination, marking the subtask as completed can avoid other The destination side repeats the same task.

Moreover, after obtaining the photos uploaded by the part of the destination, the crowdsourcing platform may further model the 3D model of the target object according to the received photos, and adjust the currently unfinished according to the completion degree of the partial modeling. Subtasks.

For example, the crowdsourcing platform can attempt to perform partial 3D modeling of the target object in real time. If the images between two non-adjacent points can be spliced to meet the modeling requirements, then all the unfinished points in the two points Photo shooting tasks can be marked as completed or canceled directly, and these points will no longer be visible in all subtasks received by the destination.

Further, after obtaining the photo uploaded by the destination end, the crowdsourcing platform may send the photo to the initiator, where the photo is used by the initiator to establish a 3D model of the target object. In another possible implementation manner of the embodiment of the present disclosure, after obtaining the photo uploaded by the destination end, the crowdsourcing platform may also establish a 3D model of the target object according to the photo, and send the 3D model to the initiator. That is to say, the 3D model establishment of the target object can be performed by the crowdsourcing platform or by the originating end. Based on this, after receiving the modeling task for the target object published by the initiator, the crowdsourcing platform may first query whether the 3D model of the target object already exists, for example, the crowdsourcing platform history has established a 3D model of the target object, or The other initiators have issued the same modeling task. If the 3D model of the target object already exists, the crowdsourcing platform can directly acquire the 3D model of the target object and return the 3D model to the initiator.

In the specific implementation, the crowdsourcing platform can implement a reward mechanism. The originating user wants to obtain a 3D model of a target object on the crowdsourcing platform, and needs to be provided to the platform at a certain cost, and the crowdsourcing platform can provide a certain reward. For example, money, points, etc. to stimulate the destination to accept and complete the task, in order to finally get the 3D model of the target object, thereby improving the efficiency of the object 3D model establishment.

Optionally, the crowdsourcing platform may also present the completion percentage to the initiator according to the number of completions of the subtasks, so that the current modeling progress is mastered. In the case of the reward mechanism, the initiator may also adjust the reward value according to the current progress. Speed up modeling.

In order to make those skilled in the art understand the technical solutions provided by the embodiments of the present disclosure, a method for establishing a three-dimensional image model of an object by using a crowdsourcing method according to an embodiment of the present disclosure is described in detail below.

As shown in FIG. 3, the implementation environment involved in the embodiments of the present disclosure may include an initiator 101, a crowdsourcing platform 102, and a destination 103. The initiator 101 may be a user's laptop, a tablet, a smart phone, etc., and the initiator 101 has a network communication function, and is connected to the crowdsourcing platform 102. Internet communication is possible. The destination end 103 can be a user's tablet, a smart phone, a smart camera, a drone, etc., and has network communication and photographing functions. The crowdsourcing platform 102 is a cloud server or server cluster. The end of the crowdsourcing platform can be used by the user of the originating end and the destination end in the form of a terminal APP. In FIG. 3, the originating end 101 is a notebook computer, and the crowdsourcing platform 102 is a server cluster, and the destination end 103 is a smart phone.

Based on the implementation environment shown in FIG. 3, a method for establishing an object model in a crowdsourcing manner according to an embodiment of the present disclosure is as shown in FIG. 4, including:

S401. The initiator 101 issues a modeling task to the crowdsourcing platform 102.

S402. The crowdsourcing platform 102 acquires the modeling task.

S403. The crowdsourcing platform 102 determines, according to the description information in the modeling task, that the initiator requests to establish a target object of the 3D model.

S404. The crowdsourcing platform 102 analyzes, according to the related information of the target object, a photo to be taken, which is required to establish a 3D model of the target object, and shooting coordinate information of the photo to be taken.

The related information may include description information in the modeling task, and may further include other related information of the target object obtained by the crowdsourcing platform 102 according to the description information.

S405. The crowdsourcing platform 102 generates a plurality of subtasks that require photographing of the target object according to the shooting coordinate information of each photo to be taken.

S406. The crowdsourcing platform 102 acquires the positioning information of the destination end in real time, and determines a destination end within a predetermined range of the shooting coordinates of the photo to be taken according to the positioning information, to obtain a first destination end set.

S407. The crowdsourcing platform 102 determines, according to the photographing device parameter of each destination end in the first destination set, the second destination set of the dispatch subtask.

S408. The crowdsourcing platform 102 assigns the plurality of subtasks to the destination end in the second destination set.

Among them, the crowdsourcing platform 102 can assign one subtask to multiple destinations at the same time. For example, the crowdsourcing platform 102 can assign all subtasks to coordinate positions that are at a distance from the target object. Each destination within a given range.

S409. The destination end 103 receives the subtask assigned by the crowdsourcing platform, and takes a photo of the target object at the shooting coordinate of the subtask according to the user operation.

For example, the destination end 103 may display the shooting coordinates of the subtask on the map for a certain subtask received, and may display the target object on the shooting interface when the current position of the destination end 103 reaches the shooting coordinate. The contour is convenient for the user to fill the contour of the target object and take a picture.

S410. The destination end 103 uploads the photographed photo to the crowdsourcing platform 102.

S411. The crowdsourcing platform 102 receives the photo uploaded by the destination end 103, and marks the corresponding subtask as completed when it is determined that the photo meets the modeling requirement.

Wherein, the modeling requirement can refer to the threshold requirement of the horizontal pixel value and the vertical pixel value of the target object in the photo described in the foregoing method embodiment. In addition, after marking a subtask as completed, you can set other destinations to be invisible to completed subtasks.

S412. The crowdsourcing platform 102 partially models the target object in real time according to the currently received photo.

S413. The crowdsourcing platform 102 adjusts the currently uncompleted subtask according to the partial modeling result.

For example, the crowdsourcing platform 102 may recalculate the uncaptured portion, the required photographs to be taken, and the photographing coordinate information of the photograph to be photographed based on the current partial modeling result, and regenerate the subtasks assigned to the destination.

S414. After the crowdsourcing platform 102 establishes the 3D model of the target object, the 3D model is sent to the initiator 101.

S415. The initiator 101 receives a 3D model of the target object sent by the crowdsourcing platform 102.

It should be noted that, for the above method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should know that the present disclosure is not described. The limit of the action sequence. In the following, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the present disclosure.

The embodiment of the present disclosure further provides a crowdsourcing platform for implementing a method for establishing an object model in a crowdsourcing manner provided by the foregoing method embodiment. As shown in FIG. 5, the crowdsourcing platform includes:

The task obtaining module 501 is configured to acquire a modeling task issued by the initiator, where the modeling task includes related information of the target object, and the modeling task is used to request to establish a 3D model of the target object;

The modeling analysis module 502 is configured to analyze, according to the related information, a photo to be photographed required to establish a 3D model of the target object, and shooting coordinate information of the photo to be photographed;

a subtask generating module 503, configured to: according to the photograph to be photographed and the photographing coordinate information, at least one subtask, the at least one subtask comprising requesting a photographing task on the target object;

a task dispatching module 504, configured to assign the at least one subtask to at least one destination;

The receiving module 505 is configured to receive a photo uploaded by the at least one destination for responding to the at least one subtask, where the photo is used to establish a 3D model of the object.

Using the above-mentioned crowdsourcing platform, after obtaining the modeling task issued by the initiator, the crowdsourcing platform can automatically analyze the 3D model needed to establish the target object according to the related information of the target object described in the modeling task. The photo, and the coordinates of the photo to be taken, and the crowdsourcing platform can divide the modeling task into a plurality of subtasks and assign them to multiple destinations based on the analysis results. In this way, for the initiator, it only needs to publish the modeling task describing the relevant information of the target object on the crowdsourcing platform, and does not need to undertake the photo shooting work required to establish the model, thereby reducing the cost and, in mass, When the destination is willing to accept and execute the task assigned by the crowdsourcing platform, the efficiency of taking the photos required for modeling will be greatly improved, thereby improving the efficiency of object modeling.

Optionally, the related information includes at least one of the following:

a picture of the target object, a name of the target object, a geographic coordinate of the target object, Height information of the target object, footprint information of the target object.

Optionally, the related information includes geographic coordinates, height information, and footprint information of the target object, and the modeling analysis module 502 is specifically configured to determine, according to the geographic coordinates, the target object on a 3D map. a location, the 3D map includes contour information of the target object; and determining, according to the contour information, the height information and the footprint information, different shooting angles required to complete a 3D model of the target object And a 3D coordinate range of a photographing position of each of the photographs to be photographed, the photographing coordinate information including the 3D coordinate range.

Optionally, the task assignment module 504 is configured to: acquire positioning information of the first destination end, and determine whether the shooting coordinate information of the first subtask and the positioning information of the first destination end meet a predetermined distance condition, where a subtask is any one of the at least one subtask; if the photographing coordinate information of the first subtask and the positioning information of the first destination end satisfy the predetermined distance condition, the The first subtask is assigned to the first destination.

Optionally, the task dispatching module 504 is configured to acquire positioning information of the second destination end and shooting device parameters of the second destination end, and determine, according to the shooting coordinate information of the second subtask and the shooting device parameter. Whether the shooting device of the second destination meets the preset shooting requirement, and the second subtask is any one of the at least one subtask; if the shooting device of the second destination meets the preset shooting requirement And the shooting information of the second subtask and the positioning information of the second destination meet a predetermined distance condition, and the second subtask is assigned to the second destination.

The shooting device parameter may include a viewing angle α of the camera and a maximum camera resolution m*n, and the shooting coordinate information in the second subtask is (x1, y1), the target object center When the point coordinates are (x2, y2), the task dispatching module 504 can calculate, according to the following formula, that the photographing device of the second destination end captures the horizontal pixel value x and the vertical pixel value y of the target object:

x=m*c/(2*tan(α/2)*√((x1–x2) ² +(y1–y2) ² ));

y=n*b/(2*tan(α/2)*√((x1–x2) ² +(y1–y2) ² ));

Where c is the width of the target object relative to the photographing coordinate direction of the second subtask, b is the height of the target object relative to the photographing coordinate direction of the second subtask, and the lateral pixel value x is determined Whether it is greater than a preset horizontal threshold, and determining whether the vertical pixel value y is greater than a preset vertical threshold; wherein, if the horizontal pixel value x is greater than the preset horizontal threshold and the vertical pixel value y is greater than the preset The longitudinal threshold determines that the photographing device of the second destination meets the preset photographing requirement.

Optionally, the subtask includes photographing guide information for guiding the destination end to reach a position specified by the photographing coordinate information, and guiding the destination end to display an outline of the photo to be photographed of the subtask .

Optionally, as shown in FIG. 6, the crowdsourcing platform further includes: a sending module 506 and a remote control module 507, where the sending module 506 is configured to send a remote control request to the destination end, and the remote control module 507 is configured to receive the module. After receiving the response sent by the destination end to confirm the permission of the remote control, the 505 remotely controls the destination end to complete the subtask received by the destination end.

Optionally, as shown in FIG. 6, the crowdsourcing platform further includes: a photo detection module 508, configured to determine whether the received photo for responding to the third subtask meets a preset condition, and the third subtask is Any one of the at least one subtasks; the task marking module 509 is configured to mark the third subtask as if the photo for responding to the third subtask meets the preset condition carry out.

Optionally, as shown in FIG. 6, the crowdsourcing platform further includes: a modeling module 510, configured to perform partial modeling of the 3D model on the target object according to the received photo. The task dispatching module 504 is further configured to adjust the currently uncompleted subtask according to the degree of completion of the partial modeling.

Optionally, the sending module 506 of the crowdsourcing platform is further configured to send the photo to the office The originating end, the photo is used by the initiator to establish a 3D model of the target object; or the sending module 506 is further configured to: establish, in the modeling module 510, the 3D of the target object according to the photo After the model, the 3D model is sent to the initiator.

It will be apparent to those skilled in the art that for the convenience and brevity of the description, only the division of each functional module described above is exemplified. In practical applications, the above functional assignments may be completed by different functional modules as needed. The internal structure of the crowdsourcing platform is divided into different functional modules to perform all or part of the functions described above. For the specific working process of the foregoing function module, refer to the corresponding process in the foregoing method embodiment, and details are not described herein again.

Thus, in a possible application scenario of the embodiment of the present disclosure, the user may wish to see images of various angles of a well-known building or any other building, landmark, etc., or the user travels locally, but because of the location limitation, only When you can see a certain angle of the building, even if the user has the software to be modeled, it takes a lot of manpower and resources to take photos in the angle and position of the required requirements, which may not be able to complete their own capabilities. At this time, the 3D model of the building to be viewed can be obtained by the crowdsourcing platform provided by the embodiment of the present disclosure, which reduces the cost and improves the modeling efficiency.

The embodiment of the present disclosure further provides a non-transitory computer readable storage medium 1 including one or more programs for performing the above method implementation. The example provides a method of crowdsourcing to build an object model.

The embodiment of the present disclosure further provides a crowdsourcing platform 2, comprising: the non-transitory computer readable storage medium 1 and one or more processors for executing the non-transitory computer readable storage medium 1 In the program.

FIG. 7 is another schematic structural diagram of a crowdsourcing platform according to an embodiment of the present disclosure. Referring to FIG. 7, a crowdsourcing platform includes a processor 701, which may be one or more, and a memory 702 for storing a processor. 701 computer program executed. The computer program stored in memory 702 can include one or more modules each corresponding to a set of instructions. Moreover, processor 702 can be configured to execute the computer program to perform the method of building an object model in a crowdsourced manner as described above.

In addition, the crowdsourcing platform can also include a power component 703 and a communication component 704 that can be configured to perform power management of the crowdsourcing platform, the communication component 704 can be configured to enable communication of the crowdsourcing platform, for example, wired Or wireless communication. In addition, the crowdsourcing platform can also include an input/output (I/O) interface 705. The crowdsourcing platform can operate based on operating systems stored in memory 702, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, and the like.

An embodiment of the present disclosure further provides an object modeling system, where the system includes a crowdsourcing platform, and an originating end and a destination end, wherein the originating end and the destination end are terminal devices registered in the crowdsourcing platform. . The system can refer to the initiator 101, the crowdsourcing platform 102, and the destination 103 in the implementation environment shown in FIG. The specific structure of the crowdsourcing platform can be referred to the crowdsourcing platform shown in FIG. 5 or FIG. 6, and details are not described herein again.

The preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings. However, the present disclosure is not limited to the specific details of the above embodiments, and various simple modifications can be made to the technical solutions of the present disclosure within the scope of the technical idea of the present disclosure. These simple variations are all within the scope of the disclosure.

It should be further noted that the specific technical features described in the above specific embodiments may be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present disclosure is applicable to various possibilities. The combination method will not be described separately.

In addition, any combination of various embodiments of the present disclosure may be made as long as it does not deviate from the idea of the present disclosure, and should also be regarded as the disclosure of the present disclosure.

Claims

A method for establishing a three-dimensional image model of an object, wherein the method is applied to a crowdsourcing platform, the method comprising:

Obtaining a modeling task issued by the initiator, where the modeling task includes related information of the target object, and the modeling task is used to request to establish a 3D model of the target object;

And analyzing, according to the related information, a photo to be photographed required to establish a 3D model of the target object, and shooting coordinate information of the photo to be photographed;

Generating at least one subtask according to the photograph to be photographed and the photographing coordinate information, the at least one subtask comprising a task of requesting photographing of the target object;

Assigning the at least one subtask to at least one destination;

Receiving a photo uploaded by the at least one destination for responding to the at least one subtask, the photo being used to establish a 3D model of the object.
The method of claim 1, wherein the related information comprises at least one of the following:

a picture of the target object, a name of the target object, a geographic coordinate of the target object, a height information of the target object, and an area information of the target object.
The method according to claim 1, wherein the related information includes geographical coordinates, height information, and footprint information of the target object, and the analyzing and establishing the 3D of the target object according to the related information The photograph to be photographed required by the model, and the photographing coordinate information of the photograph to be photographed, including:

Determining a location of the target object on a 3D map according to the geographic coordinates, the 3D map including contour information of the target object;

And determining, according to the contour information, the height information and the floor space information a 3D coordinate range of a photographing position of a different photographing angle required for the 3D model of the target object and a photographing position of each of the photographs to be photographed, the photographing coordinate information including the 3D coordinate range.
The method according to any one of claims 1 to 3, wherein the assigning the at least one subtask to at least one destination comprises:

Obtaining positioning information of the first destination end;

Determining whether the shooting coordinate information of the first subtask and the positioning information of the first destination end satisfy a predetermined distance condition, and the first subtask is any one of the at least one subtask;

And if the shooting coordinate information of the first subtask and the positioning information of the first destination end satisfy the predetermined distance condition, the first subtask is assigned to the first destination end.
The method according to any one of claims 1 to 3, wherein the assigning the at least one subtask to at least one destination comprises:

Obtaining positioning information of the second destination end and shooting device parameters of the second destination end;

Determining, according to the photographing coordinate information of the second subtask and the photographing device parameter, whether the photographing device of the second destination end satisfies a preset photographing request, and the second subtask is any one of the at least one subtask Subtasks;

And if the photographing device of the second target end meets the preset photographing requirement, and the photographing coordinate information of the second subtask and the positioning information of the second destination end satisfy a predetermined distance condition, then the second A subtask is assigned to the second destination.
The method according to claim 5, wherein said photographing device parameters comprise a field of view angle α of the camera and a maximum photographing resolution m*n, said photographing of said second subtask The target information is (x1, y1), the target object center point coordinate is (x2, y2), and the photographing coordinate information according to the second subtask and the photographing device parameter determine the photographing of the second target end Whether the device meets the preset shooting requirements, including:

Calculating the lateral pixel value x and the vertical pixel value y of the target object by the photographing device of the second destination end according to the following formula:

x=m*c/(2*tan(α/2)*√((x1–x2) 2 +(y1–y2) 2 ));

y=n*b/(2*tan(α/2)*√((x1–x2) 2 +(y1–y2) 2 ));

Where c is the width of the target object relative to the shooting coordinate direction of the second subtask, and b is the height of the target object relative to the shooting coordinate direction of the second subtask;

Determining whether the horizontal pixel value x is greater than a preset lateral threshold, and determining whether the vertical pixel value y is greater than a preset longitudinal threshold; wherein if the horizontal pixel value x is greater than the preset lateral threshold and the vertical pixel The value y is greater than the preset vertical threshold, and it is determined that the photographing device of the second destination meets the preset photographing requirement.
The method according to any one of claims 1 to 3, wherein the subtask includes photographing guide information for guiding the destination end to reach a position specified by the photographing coordinate information, and guiding the destination end An outline of the photo to be taken of the subtask is displayed.
The method according to any one of claims 1 to 3, further comprising: after the assigning the at least one subtask to the at least one destination end, further comprising:

Sending a remote control request to the destination;

After receiving the response sent by the destination end for confirming the permission of the remote control, the destination end is remotely controlled to complete the subtask received by the destination end.
Method according to any one of claims 1 to 3, characterized in that the method further include:

Determining whether the received photo for responding to the third subtask meets a preset condition, and the third subtask is any one of the at least one subtask;

If the photo for responding to the third subtask satisfies the preset condition, the third subtask is marked as completed.
The method of claim 9 wherein the method further comprises:

Partial modeling of the 3D model of the target object based on the received photo;

The currently unfinished subtasks are adjusted according to the degree of completion of the partial modeling.
The method according to any one of claims 1 to 3, wherein the method further comprises:

Sending the photo to the originating end, the photo being used by the originating end to establish a 3D model of the target object; or

Establishing a 3D model of the target object according to the photo, and transmitting the 3D model to the originating end.
A crowdsourcing platform, characterized in that

a task acquisition module, configured to acquire a modeling task issued by the initiator, where the modeling task includes related information of the target object, where the modeling task is used to request to establish a 3D model of the target object;

a modeling analysis module, configured to analyze, according to the related information, a photo to be photographed required to establish a 3D model of the target object, and shooting coordinate information of the photo to be photographed;

a subtask generating module, configured to generate at least one subtask according to the photograph to be photographed and the photographing coordinate information, where the at least one subtask includes a task of requesting photographing the target object;

a task dispatching module, configured to assign the at least one subtask to at least one destination;

And a receiving module, configured to receive a photo uploaded by the at least one destination for responding to the at least one subtask, where the photo is used to establish a 3D model of the object.
A non-transitory computer readable storage medium, comprising: one or more programs, the one or more programs for performing any one of claims 1 to 11 The method described in the item.
A crowdsourcing platform, comprising: the non-transitory computer readable storage medium of claim 11;

One or more processors for executing a program in the non-transitory computer readable storage medium.
An object modeling system, comprising: the crowdsourcing platform according to claim 11 or 13, and an originating end and a destination end, wherein the originating end and the destination end are in the The terminal device registered by the crowdsourcing platform.