CN117971361A

CN117971361A - Display processing method and device of labeling interface and electronic equipment

Info

Publication number: CN117971361A
Application number: CN202311182833.XA
Authority: CN
Inventors: 李晢曦
Original assignee: BOE Technology Group Co Ltd
Current assignee: BOE Technology Group Co Ltd
Priority date: 2023-09-13
Filing date: 2023-09-13
Publication date: 2024-05-03

Abstract

The embodiment of the disclosure provides a display processing method, a system, a device and electronic equipment of a labeling interface, belongs to the technical field of data processing, and aims to avoid labeling conflict, wherein the method comprises the following steps: responding to the opening operation of the first user on the target data, and acquiring the current state information of the target data; wherein the state information includes a first state and a second state; based on the state information, determining various display contents to be displayed and editing states corresponding to the various display contents; the display content comprises at least target data and a control for marking the target data, and the editing state is used for representing whether the display content can be edited or used; and displaying various display contents according to the editing state in the annotation interface pushed to the first user.

Description

Display processing method and device of labeling interface and electronic equipment

Technical Field

The disclosure relates to the technical field of data processing, and in particular relates to a display processing method and device of a labeling interface and electronic equipment.

Background

With the development of deep learning in recent years, great breakthroughs are made in the aspects of computer vision, natural voice processing and the like. In the deep learning field, it is generally necessary to construct training samples, and training a neural network model for a specific task by using the training samples under a certain training task.

In the process of constructing the training sample, data labeling is generally required for the constructed training sample, for example, the training sample is an image sample, and the training task is a position detection task, for example, the position of a specific object, for example, a position frame, is required to be labeled on the image sample.

With deep learning being deeply applied in various fields, the demand for data labeling is greatly increased, so that a team is generally adopted to label data, and a plurality of people can finish the data labeling of a training sample under a training task.

Disclosure of Invention

Based on the content of the background technology, the disclosure provides a display processing method and device of a labeling interface and electronic equipment.

In a first aspect of the present disclosure, a display processing method of an annotation interface is provided, the method including:

responding to the opening operation of a first user on target data, and acquiring the current state information of the target data; wherein the state information includes a first state that characterizes whether the target data is being annotated, and a second state that characterizes whether the target data has been annotated;

Based on the state information, determining various display contents to be displayed and editing states corresponding to the various display contents; wherein the plurality of display contents at least comprise the target data and a control for annotating the target data, and the editing state is used for representing whether the display contents can be edited or used;

Displaying a plurality of display contents according to the editing state in an annotation interface pushed to the first user; the same display content corresponds to different editing states in different first states, and different second states correspond to different kinds of display content.

Optionally, the responding to the opening operation of the first user to the target data obtains the state information of the target data, including:

Responding to the opening operation, triggering the calling of a first interface and the calling of a second interface, wherein the first interface is used for acquiring the first state, and the second interface is used for acquiring the second state;

And determining the state information based on the states acquired by the first interface and the second interface respectively.

Optionally, the determining, based on the state information, a plurality of display contents to be displayed and editing states corresponding to the plurality of display contents respectively includes:

determining that a plurality of display contents comprise the target data and a plurality of controls, and determining that the editing state of a plurality of controls is a state which can be used and the editing state of the target data is a state which can be edited under the condition that the first state represents that the target data is not marked and the second state represents that the target data is not marked;

And determining that various display contents comprise the target data and a plurality of controls and determining that the editing state of the plurality of controls is an unusable state and/or the editing state of the target data is an uneditable state under the condition that the first state represents that the target data is being marked and the second state represents that the target data is not marked.

Optionally, in the case that the second state characterizes that the target data has been marked, marking data of the target data is further included in the multiple display contents; wherein, in the case that the first state characterizes that the target data is not being marked, the editing state of the marking data is an editable state.

Optionally, in a case that the editing status of the plurality of display contents in the annotation interface is editable and usable, the method includes:

setting a first state of the target data as a state being marked in the process of marking the target data in the marking interface by the first user;

responding to the opening operation of a second user on the target data, and pushing an annotation interface corresponding to the second user based on the state being annotated;

and the editing state of part or all of the display content in the labeling interface corresponding to the second user is in an unavailable state so as to prohibit the second user from labeling the target data.

Optionally, after the displaying of the plurality of display contents according to the editing state in the labeling interface, the method further includes:

Under the condition that the editing state corresponding to each of the display contents is an editable state, responding to the marking operation of the first user on the target data in the marking interface, and acquiring marking data corresponding to the marking operation; wherein the annotation data comprises text data and/or graphic data;

Acquiring annotation parameters corresponding to the annotation data; the annotation parameters are used for describing the annotation data;

generating a labeling file corresponding to the target data based on the labeling parameters and the identification of the target data;

and packaging the annotation file according to a preset format, associating the packaged annotation file with the target data, and uploading the annotation file to a file storage server.

responding to the labeling operation of the first user on the target data, and acquiring corresponding labeling parameters based on the labeling data for labeling the target data;

storing the labeling parameters into a target field of the target data; the target field is used for describing the position, the attribute and the identification of the annotation data;

The target field is used for being called when the annotation interface is rendered, so that the annotation data is rendered into the annotation interface based on the annotation parameters.

Optionally, the method further comprises:

Acquiring the operation authority of the first user;

The determining, based on the state information, a plurality of display contents to be displayed and editing states corresponding to the plurality of display contents respectively includes:

Based on the operation authority and the state information, determining various display contents to be displayed and editing states corresponding to the various display contents;

And the display content with the corresponding editing state in the labeling interface is the display content with the operation authority of the first user.

Optionally, the labeling task corresponding to the target data comprises a plurality of labeling subtasks; the method further comprises the steps of:

determining a target labeling subtask corresponding to the first user in a plurality of labeling subtasks;

based on the state information, determining a plurality of display contents to be displayed and the corresponding editing states of the display contents in the labeling interface of the target data, wherein the editing states comprise:

Based on the state information and the target labeling subtask, determining various display contents to be displayed and editing states of the various display contents in a labeling interface of the target data;

and displaying the display content which corresponds to the editing state in the labeling interface and is in an editable state as the display content under the target labeling subtask.

Optionally, the determining of the labeling interface includes:

determining the annotation type corresponding to the target data;

Based on the annotation type and the state information, determining various display contents to be displayed and editing states corresponding to the various display contents;

And the display content with the corresponding editing state in the labeling interface in the editable state is the display content matched with the labeling type.

Optionally, when detecting that a fifth user performs an open operation on the target data at the same time when the first user performs the open operation on the target data, the method further includes:

if the first state of the target data is determined to be a state which is not marked, outputting the marking interface to the first user and the fifth user according to the editing states corresponding to the display contents;

Setting the editing state of the display content in the labeling interface corresponding to the target user to be an editable state for the target user which performs labeling operation in advance in the first user and the fifth user so as to allow the target user to label the target data;

And setting the editing state of the display content in the corresponding labeling interface to be an uneditable state for the user who executes the labeling operation later in the first user and the fifth user so as to prohibit the user who executes the labeling operation later from labeling the target data.

Obtaining labeling data of labeling the target users of the first user and the fifth user, wherein the labeling operation is completed on the target users;

and generating a labeling file corresponding to the target data based on the labeling data.

In a second aspect of the present disclosure, there is provided a display processing apparatus for an annotation interface, the apparatus comprising:

The state information acquisition module is used for responding to the opening operation of the first user on the target data and acquiring the current state information of the target data; wherein the state information includes a first state that characterizes whether the target data is being annotated, and a second state that characterizes whether the target data has been annotated;

The information determining module is used for determining various display contents to be displayed and editing states corresponding to the various display contents respectively based on the state information; wherein the plurality of display contents at least comprise the target data and a control for annotating the target data, and the editing state is used for representing whether the display contents can be edited or used;

The pushing module is used for displaying a plurality of display contents according to the editing state in a labeling interface pushed to the first user; the same display content corresponds to different editing states in different first states, and different second states correspond to different kinds of display content.

According to the embodiment of the first aspect of the disclosure, the current state information of the target data can be obtained in response to the opening operation of the first user on the target data; based on the state information, determining various display contents to be displayed and the corresponding editing states of the various display contents in the labeling interface of the target data; then, displaying various display contents according to the editing state in a labeling interface pushed to the first user; the state information comprises a first state of whether the target data is being marked or not and a second state of whether the target data is marked or not, and various display contents can comprise the target data and a control for marking the target data.

When a first user opens target data, a first state of whether the target data is marked by other users and a second state of whether the target data is marked by other users can be determined, display content displayed for the first user is determined together according to the first state and the second state, whether the display content can be edited or not is determined together according to the first state and the second state, whether the editing states of various display contents are marked currently or are associated with the target data or not in a marking interface pushed to the first user or not is determined currently or not, and therefore when the team marks the data, the current first user can be controlled to mark the target data according to the condition that the target data is marked by other users, marking of the same target data by a plurality of users at the same time can be avoided, efficiency of marking the data by the team can be improved, repeated marking of the same target data by the plurality of users can be avoided, and effectiveness and accuracy of marking the data can be improved.

The embodiment of the disclosure also discloses an electronic device, which comprises: comprising a memory, a processor and a computer program stored in the memory and executable on the processor, which when executed implements a method for processing the display of an annotation interface as described in the embodiment of the first aspect.

In an embodiment of the disclosure, a computer readable storage medium is also disclosed, where a computer program stored in the storage medium causes a processor to execute the display processing method of the labeling interface according to the first aspect of the disclosure.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings that are needed in the description of the embodiments of the present disclosure will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present disclosure, and other drawings may be obtained according to these drawings without inventive effort to a person of ordinary skill in the art.

FIG. 1 is a schematic view of an implementation environment of a display processing method of a label interface in an embodiment of the disclosure;

FIG. 2a is a schematic step flow diagram of a display processing method of a label interface in an embodiment of the disclosure;

FIG. 2b is a schematic diagram of a display processing method of a label interface in an embodiment of the disclosure;

FIG. 3 illustrates an interface schematic for creating a dataset in an embodiment of the present disclosure;

FIG. 4 illustrates an interface diagram for creating annotation tasks for datasets in an embodiment of the present disclosure;

FIG. 5 illustrates an interface change diagram for rights management of a data set in an embodiment of the present disclosure;

FIG. 6 illustrates an interface diagram for selecting annotation tasks in an embodiment of the present disclosure;

FIG. 7 illustrates an interface diagram of one annotation interface pushed to a first user in an embodiment of the present disclosure;

FIG. 8 shows a flow diagram of steps for the generation of a marked file in an embodiment of the present disclosure;

fig. 9 is a schematic structural diagram of a display processing device of the label interface in the embodiment of the present disclosure.

Detailed Description

In order that the above-recited objects, features and advantages of the present disclosure will become more apparent, a more particular description of embodiments of the disclosure will be rendered by reference to the appended drawings, which together with the appended drawings illustrate embodiments of the disclosure, wherein it is obvious that the described embodiments are some, but not all, of the embodiments of the disclosure. All other embodiments, which can be made by one of ordinary skill in the art without inventive effort, based on the embodiments in this disclosure are intended to be within the scope of this disclosure.

In the related art, in a mode of marking data by adopting a team, a plurality of members in the team can jointly mark the data of a plurality of data under the same marking task. However, when team is adopted for data annotation, annotation conflicts exist, which are embodied as follows:

If multiple persons label the same data at the same time, there may be labeling conflict, at this time, the stored labeling data is only the person who finishes labeling at the last time, but the data of the previous user for labeling will be lost, for example, the user A and the user B label one data at the same time, the user A labels the data first, and the user B labels the data later, at this time, the system will only store the data of the user B, but the data of the user A label will be lost, not only resulting in the waste of the labeled manpower resources, but also reducing the labeling efficiency of the whole data set due to repeated labeling of the same data.

In view of this, the disclosure proposes a display processing method for an annotation interface, which achieves the purpose of avoiding annotation conflicts by processing display contents to be displayed on the annotation interface pushed to a user. Specifically, two states, namely a first state and a second state, can be set for target data to be marked, the first state represents whether the target data is marked, the second state represents whether the target data is marked, according to the two states, not only display content to be displayed on a marking interface is determined, but also an editing state of the display content on the marking interface is determined, and in the marking interface pushed to a user, the display content is displayed according to the editing state corresponding to the display content. Therefore, when the target data is marked by other users, the corresponding editing state in the display interface is set to control the marking of the target data by the current first user, so that the marking of the same target data by a plurality of users at the same time can be avoided, the efficiency of marking the data by a team can be improved, the repeated marking of the same target data by a plurality of users can be avoided, and the effectiveness and the accuracy of the data marking can be improved.

Referring to fig. 1, an implementation environment schematic diagram of a display processing method of an labeling interface of the present disclosure is shown, where, as shown in fig. 1, a platform includes multiple data sets, each data set includes multiple data to be labeled, and different data sets may correspond to the same labeling task or different labeling tasks; wherein, also include a plurality of teams, each team includes a plurality of users, different teams can be used for accomplishing different labeling tasks.

The plurality of users in the same team are all used for completing data annotation under the same annotation task, so that the plurality of users in the same team can carry out data annotation on the plurality of data under the same data set. Different users can use clients running on different electronic devices, log in the clients, and enter the platform to label data in corresponding data sets in the platform.

The client in this embodiment may be a web client, and the platform may be a web server connected to the web client.

Referring to fig. 2a and 2b, fig. 2a shows a step flow diagram of a display processing method of a labeling interface, and fig. 2b shows a frame diagram of a display processing method of a labeling interface; the method can be applied to the platform and the client shown in the figure 1, and the platform and the client can execute respective corresponding operations in the method so as to jointly complete the display processing of the labeling interface; as shown in fig. 2a and 2b, the method specifically comprises the following steps:

step S201: responding to the opening operation of the first user on the target data, and acquiring the current state information of the target data;

Wherein the state information includes a first state of whether the target data is being marked and a second state of whether the target data has been marked.

Step S202: based on the state information, determining various display contents to be displayed and editing states corresponding to the various display contents;

The display content comprises at least target data and a control for labeling the target data, and the editing state is used for representing whether the display content can be edited or used.

In this embodiment, the first user may be any one of the users in the team, and the target data may be image data or text data. The target data may be data in a data set to be marked, and the data set to be marked may be uploaded to the platform in advance.

Specifically, a dataset may be created first, and referring to FIG. 3, an interface schematic for creating a dataset is shown. As shown in fig. 3, each dataset may include a plurality of data, which may be of the same type, such as data of both image types, or data of both text types; or the dataset may comprise a plurality of different types of data, such as data comprising an image type as well as data comprising a video type. After the data set is created, the data set may be uploaded to a platform and associated with a corresponding labeling task, and referring to fig. 4, an interface schematic diagram of creating the labeling task for the data set is shown, as shown in fig. 4, a task creation interface may be popped up in response to an operation of creating the labeling task by a user, and a plurality of setting fields may be included in the task creation interface, where different setting fields are used for setting different parameters of the labeling task, such as a task name setting field, a data set selection field, a data type setting field, and a labeling type setting field. The labeling type setting column may include, as shown in fig. 4, options such as image classification, object detection, image segmentation, and OCR (Optical Character Recognition ), where, for image classification, it refers to a category in which an image needs to be labeled; for target detection, it means that the position of an object in an image needs to be detected, for example, the position of a specific object in the image is marked; for image segmentation, it means that two different objects in the image need to be divided; for OCR, it refers to text content that needs to be annotated to a certain region in an image.

After the labeling task and the creation of the dataset are completed, it may be determined whether the dataset is set to be public, i.e., whether the dataset is visible to the user, in general, the published dataset may be labeled by a team, the non-published dataset is visible only to the user uploading the dataset, and the remaining users are not visible.

After uploading the data set to the platform, the labeling state of the data set may be marked, in general, for the data set to be labeled, the labeling state may be set to an unlabeled state, and for the data set that has been labeled, the labeling state may be set to a labeled state, where for the data set in the unlabeled state, the labeling needs to be labeled by a team.

It will be appreciated that the target data described above is data in a data set in an unlabeled state.

After the data set is uploaded to the platform, the labeling authority of the data set may be set, where the labeling authority may represent a team allowed to label the data set, for example, referring to fig. 5, an interface schematic diagram for performing authority management on the data set is shown, as shown in fig. 5, including an ID of the data set, that is, an identification of the data set, in user group management, a user group management on the left side may be clicked, so as to pop up an interface on the right side, and in the interface on the right side, including a button for creating a user group, and a team list, clicking an edit control in the team list, as shown in fig. 5, clicking an edit control of a team with an ID of 230 may pop up an interface under fig. 5, so as to pull a team member for the data set. The team name corresponding to each data set can be set to indicate that the corresponding team has authority to mark the data in the data set, and other teams have no authority to mark the data in the data set, so that the visible range of the data set is accurate to a user, and certain data isolation is realized.

The first user may refer to a user having a labeling authority for a data set where the target data is located, the first user may log in to the platform, first the first user selects a labeling task that needs to be labeled by the first user, the platform may respond to a selection operation of the first user on the labeling task, and display a task list, and referring to fig. 6, an interface schematic diagram when the labeling task is selected is shown, where the task list may include a plurality of labeling tasks and team names participating in labeling. Then, in response to the first user's selection operation on the labeling task, a data list corresponding to the selected labeling task may be displayed, where a plurality of data included in the data set is listed, as shown in fig. 6, data with IDs 137, 252, 125, and 123 are listed, and an identification of each data in the data set may be included in the data list.

The first user can select a certain target data in the data list to be marked, and specifically, the platform can start to provide a marking interface for the first user in response to the opening operation of the first user on the target data.

In this embodiment, in order to avoid labeling conflicts, that is, to avoid that a plurality of users label the same labeling data at the same time, each data in the dataset adopts the labeling condition of the characterization data in two states, that is, includes a first state and a second state. The first state is used for representing whether the target data is being marked, namely the first state can indicate whether other users are marking the target data at the moment when the first user opens the target data, and in practice, if the first user opens the target data to be marked, and if the other users are marking the target data, in order to avoid marking conflict, the first user cannot mark the target data.

The second state is used to characterize whether the target data is marked, that is, the second state may indicate whether the target data has a mark at the moment when the first user opens the target data, for example, other users have completed marking the target data, in practice, if the first user opens the target data to prepare the mark, and if other users have completed marking the target data, the first user may modify the mark of the target data, or the first user may not allow marking the target data any more.

The platform may maintain a first state and a second state of each data in the dataset, specifically, if the data is marked by a user, the platform sets the first state of the data to be a marked state, and if the data is not marked by any user, the platform may set the first state of the data to be a non-marked state. If the data is marked, i.e. marked data has been generated, the platform sets the second state of the data to a marked state, and if the data is not marked with marked data, the platform sets the second state of the data to an unmarked state.

It will be appreciated that the first state and the second state are two independent states for characterising the labelling state of the data at different stages, the first state being related to the labelling behaviour of the user on the data and the second state being related to whether the data is produced with labelling data.

After determining the first state and the second state of the target data, state information of the target data may be formed, where the state information includes the first state and the second state of the target data at present, specifically, the first state in the state information may exist in the form of a state code, and the second state may also exist in the form of a corresponding state code.

Based on the state information, it can be determined whether the target data is being marked by other users and whether the target data has been marked, and the marking ranges of the first user to mark are different in different first states and second states of the target data.

The display content is content to be displayed in the labeling interface, that is, content to be visible to the first user, and in one case, no matter what the first state and the second state of the target data are, the target data and the control for labeling are visible to the first user, the editing states of the target data and the control can be determined according to whether the target data is being labeled and whether the target data already has labeling data. For example, when the first state indicates that other users are already labeling the target data, the editing state of the control may be an unusable state, so that the first user cannot label the target data by using the control, and then label conflict is avoided.

For another example, when the first state indicates that no other user is annotating the target data, the edit state of the control may be a state that can be used, such that the first user can annotate the target data with the control.

For the second state, the second state may be used to determine whether the display content is to display the annotation data, and if the target data already has the annotation data, the annotation data may be included in the plurality of display contents. If the target data does not have the labeling data, the labeling data may not be included in the various display contents.

The second state may jointly determine an editing state of each display content, for example, the second state characterizes that the target data already has labeling data, and the editing state of the control may be set to an unusable state to avoid repeated labeling of the labeling data.

Step S203: displaying various display contents according to the editing state in a labeling interface pushed to a first user;

the same display content corresponds to different editing states in different first states, and different second states correspond to different kinds of display content.

In this embodiment, when a plurality of display contents and an editing state corresponding to each display content are determined, an annotation interface may be pushed to the first user, where each display content displayed in the annotation interface corresponds to the editing state determined according to the state information.

When the editing state of the display content is a state which can not be used, the color of the display content in the annotation interface can be gray, for example, the control is gray; for the client, the editing state is an unusable control, and the response to the triggering operation on the marking interface is forbidden, so that the first user cannot mark the target data by using the control.

In one example, when the edit status of the control is an unusable state, in the annotation interface of the first user, the control may be hidden in the annotation interface in accordance with its unusable state, i.e., the control is not visible to the first user. Referring to FIG. 7, an interface diagram of a labeling interface pushed to a first user is shown, as shown in FIG. 7, with a first state of target data representing what is being labeled as an example, with content displayed in the labeling interface including target data, i.e., the bicycle image of FIG. 7; wherein the control for labeling appears gray or is hidden in the target interface.

Generally, referring to fig. 2b, the web server is the platform described above, where a dataset and a team corresponding to the dataset may be created first, and multiple users in the team may be created, when a user opens target data to be marked from the web front, the web front may determine a first state and a second state of the target data from the web server, then the web server may determine multiple display contents to be displayed and an edit state corresponding to each display content according to the first state and the second state, so as to form a marking interface, push the interface to the web front, display the marking interface on the web front, and may transmit the marking data to the web server in response to the marking of the target data by the user, and the web server may further form a marking file named by an identifier of the target data, and then upload the marking file to the file storage server.

By adopting the technical scheme of the embodiment of the disclosure, when the first user opens the target data, the display content displayed to the first user and the editing state of whether the display content can be edited or not can be determined together according to the first state and the second state of the target data, so that in the labeling interface pushed to the first user, the editing states of various display contents are related to whether the target data is currently being labeled or whether the target data is currently labeled or not. Therefore, when the team marks the data, the current first user can be controlled to mark the target data according to the condition that the target data is marked by other users, so that the plurality of users can be prevented from marking the same target data at the same time, the efficiency of marking the data by the team is improved, the repeated marking of the plurality of users to the same target data can be avoided, and the effectiveness and the accuracy of marking the data are improved.

In an optional example, according to a first state and a second state of the target data, setting corresponding editing states in the display, so as to control a labeling range of the target data by a current first user, determining that multiple display contents comprise the target data and multiple controls when the first state represents that the target data is not being labeled and the second state represents that the target data is not being labeled, and determining that the editing states of the multiple controls are available and the editing states of the target data are available;

In the case where the first state characterization object data is being annotated and the second state characterization object data is not annotated, determining that the plurality of display contents includes the object data and the plurality of controls, and determining that an edit state of the plurality of controls is an unusable state and/or that an edit state of the object data is an uneditable state.

Specifically, the condition that the first state represents that the target data is not being marked and the second state represents that the target data is not marked indicates that when the first user opens the target data, no other user marks the target data, and the target data does not have marking data yet, the first user is required to mark the target data, the target data is required to be displayed in a marking interface, and a control capable of marking the target data is required to be displayed, the editing state of the control is required to be in a state capable of being used, and the target data is required to be in a state capable of being edited.

The first state represents the condition that the target data is being marked and the second state represents the condition that the target data is not marked, namely when the first user opens the target data, other users mark the target data, and the target data does not have marking data yet, the first user does not need to mark the target data, the target data and the control can be displayed in the marking interface, but the editing state of the control needs to be in a state that the control cannot be used, or the target data is in a state that the target data cannot be edited; or both in a state that cannot be edited. In this case, since the first user is not required to annotate the target data, the annotation of the first user can be prohibited as long as at least one of the control and the target data is in a state where it cannot be edited. In practice, it may be determined which of the control and target data is in a state that cannot be edited, depending on the labeling task.

Wherein, the target data being in a state that it cannot be edited may refer to: the display window of the target data in the labeling interface is locked, and corresponding labeling data on the control cannot be labeled in the display window. For example, taking the target data as a picture or a video as an example, if the labeling task is a task such as image segmentation, target detection, etc., the target data is locked, so that the target data cannot be operated to label a position frame, a segmentation frame, etc. thereon. As another example, taking the target data as a picture or a video as an example, if the labeling task is an OCR task, the target data may be locked, so that the labeling content cannot be associated with the image area in the target data, and thus the text of the image area on the target data cannot be labeled.

For the labeling task of the image classification task, the control can be set to be in an unusable state under the condition that the first user needs to be forbidden for labeling.

In yet another alternative example, in the case that the second state characterizes that the target data has been annotated, annotation data of the target data is further included in the plurality of display contents; wherein, in the case that the first state characterizes the target data not being marked, the editing state of the marked data is an editable state.

In this embodiment, if the target data already has the labeling data, the labeling data may be displayed on the labeling interface as display content. In this case, if no other user marks the target data, in an example, the first user may be allowed to modify the mark data in the mark interface, and on the basis that the target data and the control are in an editable state, the editing state of the mark data may also be an editable state.

Of course, in another alternative example, if the target data already has the labeling data, and if no other user labels the target data, in one example, if the first user does not have the modification authority for the labeling data, the labeling data may be displayed in the labeling interface, but the editing state of the labeling data is in a non-editable state, and the target data and the control are both in non-editable states, so as to avoid the waste of human resources caused by repeated labeling of the labeled data.

In an optional example, if the editing state of each display content in the labeling interface pushed to the first user is in a state that can be edited, the first user may label the target data in the labeling interface, and if in this process, it is detected that the second user performs the opening operation on the target data, the labeling interface whose editing state is unusable may be output to the second user.

Specifically, in the case that the editing states of the plurality of display contents in the labeling interface are editable and usable states, setting the first state of the target data to be the state being labeled in the process of labeling the target data in the labeling interface by the first user; responding to the opening operation of the second user on the target data, and pushing an annotation interface corresponding to the second user based on the state being annotated; the editing state of part or all of the display content in the labeling interface corresponding to the second user is in an unusable state so as to prohibit the second user from labeling the target data.

Specifically, since the platform can maintain the first state and the second state of the target data at the same time, when the first user starts to annotate the target data, the first state of the target data can be set to be the state being annotated. In an example, when a labeling interface capable of labeling is pushed to a first user, a first state of target data is set to be a labeling state, so that when the first user opens the target data, the first user can perform space occupation control on the target data, that is, whether the first user starts labeling or not, the target data can be locked to the first user for labeling. In another example, when the client where the first user is located detects that the first user starts to perform the labeling operation, the labeling operation may be fed back to the platform, so that the platform sets the first state of the target data to be a labeling state according to the feedback of the client, and in a process that the first user only opens the target data and does not perform the labeling operation for a long time, other users may be allowed to open the target data and perform labeling on the target data, thereby improving labeling efficiency.

After the first state of the target data is updated to be the state being marked, if the platform detects that the second user performs the opening operation on the target data, the platform pushes the editing state of part or all of the display content in the corresponding marking interface to the second user to be in an unusable state, so that the second user cannot mark the target data in the marking interface. Specifically, the partial display content may refer to that the target data is in a state that cannot be edited, or may refer to that the plurality of labeled controls are in a state that cannot be edited; the full display may then include the target data and the full controls.

Illustratively, the first user starts annotating the target data on the client, and all display content displayed on the annotation interface is editable, e.g., the controls are colored and can be triggered; while the second user can open the target data, the entire display content displayed on the annotation interface is not editable, e.g., the control is gray and cannot be triggered.

By adopting the embodiment, other users can be prevented from participating in the labeling of the target data in the process that the first user is labeling the target data, so that labeling conflict is avoided.

In an alternative example, if at the same time, both users perform an open operation on the target data, the first state and the second state of the target data are the same for both users, such as a state being marked or a state not being marked; if the first state is just the state which is not being marked, the two users can mark the target data at the same time, in order to avoid marking conflict, the display content on the marking interface of one user can be set to be in a non-editable state immediately when the other user marks first.

Specifically, when the first user performs the opening operation on the target data at the same time, and when the fifth user performs the opening operation on the target data, if the first state of the target data is determined to be a state which is not being marked, marking interfaces are output to the first user and the fifth user according to editing states corresponding to various display contents;

the method comprises the steps of setting an editing state of display content in a labeling interface corresponding to a target user in a first user and a fifth user to be an editable state for the target user which performs labeling operation in advance so as to allow the target user to label target data;

And setting the editing state of the display content in the corresponding labeling interface to be an uneditable state for the user who performs the labeling operation later in the first user and the fifth user so as to prohibit the user who performs the labeling operation later from labeling the target data.

In this embodiment, the first user and the fifth user may be two different users in the team, in some cases, the first user and the fifth user may perform an opening operation of opening the target data at the same time, and if the target data is not marked by other users at the time, the first user and the fifth user may be allowed to mark the target data at the same time. Specifically, according to the procedure of the above embodiment, the annotation interface may be pushed to the first user and the fifth user, the display content included in the pushed annotation interface is the same, and the editing state of the display content may also be the same, that is, the first user and the fifth user are allowed to annotate the target data.

In specific implementation, the client side where the first user and the fifth user are located can detect the labeling operation of the first user and the fifth user, when the respective client sides detect the labeling operation, the moment of the labeling operation can be fed back to the platform, the platform can determine the target user who executes the labeling operation in advance according to the moment of the labeling operation fed back by the two client sides, so that an operation prohibiting instruction is sent to the other user who executes the labeling operation later, and when the client side where the user who executes the labeling operation later receives the operation prohibiting instruction, the editing state of the display content on the labeling interface of the user can be set to be in a non-editable state. Specifically, the target data may be set to a non-editable state, or the control may be set to a non-editable state.

The first user and the fifth user open the target data at the same time, and the obtained labeling interfaces are all interfaces capable of labeling; if the first user performs the labeling operation before the fifth user, the platform obtains the situation and sends an operation prohibiting instruction to the client where the fifth user is located, and the control on the labeling interface corresponding to the fifth user is changed from color to gray and is in an unusable state, so that the fifth user cannot label, the control on the labeling interface corresponding to the first user is still color, and the first user can continue labeling.

In this embodiment, the labeling operation may include a triggering operation on the control, such as clicking on the control, double clicking on the control.

According to the technical scheme of the selectable example, aiming at the opening operation of a plurality of users on target data at the same moment, the same marking interface can be fed back to the plurality of users, and then the user who performs marking operation first is allowed to mark the target data according to the time sequence of marking operation of the plurality of users, and the user who performs marking operation later is forbidden to mark the target data, so that marking conflict of marking at the moment when the plurality of users agree with each other is avoided, the user who performs marking operation later can not perform marking at the next step any more, and marking is performed on other target data in turn, so that marking conflict can be avoided more reliably, and marking efficiency is improved.

In an alternative example, if at the same time, both users perform an open operation on the target data, the first state and the second state of the target data are the same for both users, such as a state being marked or a state not being marked; if the first state is just the state which is not being marked, the two users can mark the target data at the same time, in this case, in order to avoid repeated storage of the marked data, when one of the users finishes marking, the marked data of the user who finishes marking after finishing marking can be stored.

When the first user executes the opening operation on the target data, if the first state of the target data is determined to be a state which is not marked when the fifth user executes the opening operation on the target data, marking interfaces are output to the first user and the fifth user according to editing states corresponding to various display contents;

the method comprises the steps of obtaining labeling data of a target user for labeling a target user after labeling operation is completed in a first user and a fifth user; and generating a labeling file corresponding to the target data based on the labeling data.

As described above, when the first user and the fifth user open the target data at the same time, the labeling interface may be pushed to the first user and the fifth user according to the procedure of the above embodiment, the display content included in the pushed labeling interface may be the same, and the editing state of the display content may also be the same, so that the first user and the fifth user are allowed to label the target data.

In this optional example, no matter which user performs the labeling operation first, the first user and the fifth user can finish labeling the target data, then the labeling completion operation of the first user and the fifth user on the labeling data can be detected, and the labeling completion operation indicates that the user completes labeling of the target data, and the labeling data needs to be saved. In practice, labeling data of labels performed by users who finish labeling operations later in the first user and the fifth user may be saved.

In the specific implementation, the client side where the first user and the fifth user are located can respectively respond to the labeling completion operation to obtain labeling data of labeling performed by the first user and labeling data of labeling performed by the second user, the two labeling data are fed back to the platform, and the platform can store the labeling data with later completion time according to the completion time carried by the two labeling data. Or the platform can cover the label data received later and the label data received before, so that the platform can store the label data corresponding to one piece of target data, and the problem of label conflict caused by the fact that one piece of target data corresponds to a plurality of pieces of label data is avoided.

Because the platform needs to maintain the first state and the second state of each data in the dataset, in an alternative example, the platform may provide two interfaces for the client to obtain the first state and the second state, and the first state and the second state for maintaining each target data to be marked. Specifically, the two interfaces comprise a first interface and a second interface, wherein the first interface is used for being associated with a first state of target data, and can acquire the first state of the target data and maintain the first state of the target data; similarly, the second interface is configured to associate with the second state of the target data, and may acquire the second state of the target data and maintain the second state of the target data.

Correspondingly, when the state information of the target data is acquired in response to the opening operation of the first user on the target data, the calling of the first interface and the calling of the second interface can be triggered in response to the opening operation, wherein the first interface is used for acquiring a first state, and the second interface is used for acquiring a second state; the state information is determined based on the states acquired by the first interface and the second interface, respectively.

In a specific implementation, the platform may set up two APIs, ifAnnotating and annotate, where ifAnnotating is a first interface, annotate is a second interface, and any user opens the target data, triggers ifAnnotating the first interface, and sets the first state of the target data to true, that is, the state being marked by other users, when the user performs the marking operation on the target data. When other users reopen the target data, the first interface is triggered ifAnnotating, and when the first state is true, the labeling interface of the other users can display prompt information to prompt that the current target data is being labeled by the other users.

When any user opens the target data, the first interface is triggered ifAnnotating, and if the first state of the target data acquired by the first interface is false, it is indicated that the user can mark the target data.

Thus, the first state acquired by the first interface may carry a state code, such as true or false, of the target data in the first state.

When the first user marks the target data, the first interface can be called in response to the marking completion operation, the first state of the target data is set to be a state which is not marked, namely the first state is restored to false, and meanwhile, the second interface is called, the second state of the target data is set to be 1, and the marked state is indicated; of course, before the target data is not marked, the second state thereof is set to 0, which indicates that the target data is not marked, i.e., the marked data is not formed.

Thus, for a target data, when the target data is not marked, the first user who opens the target data has marking authority on the target data, and when any other member in the team opens the target data, the first state of the target data is seen to be in true state, and the marking interface provided for the target data cannot support the marking of the target data.

By adopting the technical scheme of the embodiment, the platform can independently maintain two states (the first state and the second state) of the target data through two more independent interfaces, so that the labeling process of the target data is independently maintained, and the occurrence of conflict situations that a plurality of users label the target data simultaneously in the labeling process of the target data is avoided.

In practice, in the process of marking data by a team, the level of the data marking among users in the team is uneven, the marked operations are different, if users are skilled in marking the data, the accuracy of the marking is high, and if users are not too high, the accuracy of the marking is not too high; for another example, some users are good at labeling of graphics classes, while some users are good at labeling of words classes; in order to fully adapt to the operation level of the user in the team, when the corresponding annotation interface is pushed to the user, the display content and the control in the annotation interface can be personalized set according to the operation level of the user.

For example, users with different operation authorities can execute annotation operations with different grades, for example, a user with higher authority can modify annotations of other users, while a user with lower authority cannot modify annotations of users; as another example, users with different tastes can label different types of data in the same labeling task, so that a plurality of users can respectively finish a plurality of different types of labeling data in the same labeling task, and the labeling task of the same target data can be split into a plurality of users, so that the data labeling amount born by each user is reduced, and the efficiency is improved.

Accordingly, in an alternative example, a plurality of display contents to be displayed in the annotation interface corresponding to the first user, and an editing state corresponding to each display content may be determined according to the operation authority of the first user.

Correspondingly, the operation authority of the first user can be obtained, and when various display contents to be displayed and editing states corresponding to the various display contents are determined based on the state information, the various display contents to be displayed and the editing states corresponding to the various display contents can be determined based on the operation authority and the state information;

The display content with the corresponding editing state in the editable state in the labeling interface is the display content with the operation authority of the first user.

In this embodiment, the operation authority of the first user may be used to determine whether the first user has the marking authority for the target data and the modification authority for the marking of the target data; when the first user has the labeling authority to the target data, the display content is determined to comprise the target data and the control for labeling, wherein when the target data is determined not to be labeled by other users, the editing states of the target data and the control for labeling can be determined to be editable states.

When the first user does not have the labeling authority for the target data, on one hand, the display content can be determined to comprise the target data and the control for labeling, but the editing states of the target data and the control for labeling are non-editable states; on the other hand, it may be directly determined that the display content does not include the target data, and thus, the first user may be prohibited from annotating the target data through the annotation interface.

When the first user has modification authority to the labeling data, if the second state characterizes that the target data is labeled, it can be determined that the display content includes the target data, the control and the labeling data, and then, according to the first state, the editing states of the target data, the labeling data and the control are set.

By adopting the implementation mode, the operation authority of the user on the data can be combined with the labeling condition of the data so as to jointly determine various display contents in the labeling interface and the editing state of the display contents, thereby being capable of controlling the labeling interface pushed to the user more flexibly.

In yet another alternative example, multiple display contents to be displayed in the labeling interface corresponding to the first user, and the editing state corresponding to each display content may also be determined according to the labeling task to be completed by the first user.

Accordingly, a target labeling subtask corresponding to the first user can be determined in various labeling subtasks; when determining multiple display contents to be displayed and the corresponding editing states of the multiple display contents in the labeling interface of the target data based on the state information, determining the multiple display contents to be displayed and the corresponding editing states of the multiple display contents in the labeling interface of the target data based on the state information and the target labeling subtasks;

The display content of the corresponding editing state in the labeling interface in the editable state is the display content under the subtask of the target labeling.

In this optional example, the labeling task of the data may be decomposed into multiple labeling subtasks, for example, in the target detection task, the object to be detected in the image needs to be labeled, and the position where the object to be detected is located, then the labeling object to be detected may be one labeling subtask, and the labeling position may be one labeling subtask; for another example, in the OCR labeling task, an area to be identified in the image needs to be labeled, and text content in the area to be identified needs to be labeled, so that the labeling of the area to be identified may be referred to as a labeling subtask, and the labeling of the text content may be referred to as a further labeling subtask.

In this embodiment, in the case where the labeling task of the target data is split into multiple labeling sub-tasks, users in the team may also be divided according to the labeling sub-tasks, for example, the users in the team are divided into multiple groups, the users in the same group execute the same labeling sub-task, and the users in different groups execute different labeling sub-tasks.

The controls to be used in different labeling subtasks are different, for example, in text labeling, the controls to be used can be a text input control and a text input column, and in position labeling, the controls to be used can be controls related to drawing. Accordingly, in this example, various display contents to be displayed may be determined based on the state information first, and then, editing states may be set for the various display contents based on the target labeling subtasks to be executed by the first user and the state information, and in particular, the target labeling subtasks may be used to set the editing states of the plurality of controls for labeling.

For example, the target data corresponds to an OCR labeling task, and the target labeling subtask corresponding to the first user is labeling text content in the area to be identified, so that among a plurality of controls of the labeling interface, the controls belonging to the text input control and the text input field are in a state that can be used, and the controls related to drawing are in a state that cannot be used.

By adopting the selectable example, the annotation interface can be different according to different annotation tasks born by the user, so that effective definition of the annotation task of the user can be achieved through setting display content and editing states of the display content in the annotation interface, and invalid annotation is avoided.

In yet another embodiment, the display content in the labeling interface may also relate to a labeling task corresponding to the target data, and the labeling tasks may be different, where the controls for labeling displayed on the labeling interface may be different, so that the labeling type corresponding to the target data may be determined; correspondingly, when determining a plurality of display contents to be displayed and the corresponding editing states of the plurality of display contents in the labeling interface of the target data based on the state information, determining the plurality of display contents to be displayed and the corresponding editing states of the plurality of display contents based on the labeling types and the state information;

The display content with the corresponding editing state in the labeling interface in the editable state is the display content matched with the labeling type.

Specifically, different data sets correspond to different labeling tasks, in practice, matching labeling interfaces need to be set for the different labeling tasks, in this optional example, when the labeling interfaces are pushed to users, display contents to be displayed in the labeling interfaces can be selected according to the types of the labeling tasks corresponding to target data, in particular, a control to be displayed in the labeling interfaces and used for labeling can be selected from a preset control database, and the selected control is the labeling task used for completing the target data; then, the editing state of each display content is determined according to the state information.

All controls can be stored in a preset control database, namely controls required for completing different labeling tasks can be stored, and when target data is labeled, only the controls are selected from the control library, then the controls are associated and rendered to a labeling interface, and one labeling interface is not required to be created for each labeling task in advance. Therefore, the labeling interface can be matched with the labeling task corresponding to the target data, and interface resources can be saved.

Next, processing of the labeling data obtained after the labeling is completed will be described.

In one embodiment, in order to enable the annotation data to be associated with the target data, the annotation data of the target data can be easily found, and can be stored as a single file, the annotation data is trained by a model of a later period of a user, after the annotation data is obtained, an annotation file conforming to a preset format is generated according to the identification of the target data and the annotation data, and then the annotation file is uploaded to a file storage server stored by the user for later algorithm team invocation.

Referring to fig. 8, a flowchart illustrating steps of generating a markup document is shown, and as shown in fig. 8, the method specifically may include the following steps:

Step S301: under the condition that the editing state corresponding to each of the plurality of display contents is an editable state, responding to the marking operation of the first user on the target data in the marking interface, and acquiring marking data corresponding to the marking operation; the annotation data comprises text data and/or graphic data;

Step S302: acquiring marking parameters corresponding to marking data; the marking parameters are used for describing marking data;

step S303: generating a labeling file corresponding to the target data based on the labeling parameters and the identification of the target data;

step S304: and encapsulating the annotation file according to a preset format, and uploading the encapsulated annotation file to a file storage server after associating the encapsulated annotation file with the target data.

The obtained labeling parameters may be different according to different labeling data, and if the labeling data is a labeling frame for labeling the position, the labeling parameters may include position coordinates of the labeling frame in the image, for example, position coordinates of four vertices, and in some cases, may also include a label of the labeling frame, where the label is used to indicate a category of an object in the labeling frame, and so on.

Specifically, when the client where the first user is located detects that the labeling is completed, the labeling parameters corresponding to the labeling data can be obtained, then the labeling parameters and the identification of the target data are transmitted to the web end, such as the web server, the web end firstly arranges the labeling parameters into a unified format, then encapsulates the unified format into a labeling file according to the JSON message format, takes the identification of the target data as the file name of the labeling file, then uploads the labeling file to the file storage server, the file storage server can store the labeling file in a naming mode of file name txt, and the file storage server can independently store the labeling file and the target data. Because the name of the annotation file is the same as the name of the target data, the suffix of the file is different, and the user can conveniently search the annotation file of certain target data later.

The algorithm development team member can directly go to the file server to read the corresponding annotation file.

The preset format may be a JSON (JavaScript Object Notation, JS object numbered musical notation) message body format, specifically, after the labeling data is packaged into a file according to the JSON message body format, the identifier of the target data is used as the naming of the file, and because the JSON message body format is used as the packaged file, a lightweight data exchange format is adopted between the file server and the web end, so that the data transmission efficiency is improved.

In some alternative examples, as described above, where the first user opens the target data and the second state of the target data is already annotated, the annotation data of the target data may be displayed as display content in the annotation interface regardless of whether the first state indicates that the target data is being annotated. The labeling data can be stored locally, namely locally at the client, or stored in a target field corresponding to the target data after the labeling is completed, so that when the target data is acquired and displayed, the corresponding labeling data can be directly read from the target field of the target data for display, and the display efficiency is improved.

Specifically, in response to the labeling operation of the first user on the target data, corresponding labeling parameters can be obtained based on the labeling data for labeling the target data; storing the labeling parameters into a target field of target data; the target field is used for describing the position, the attribute and the identification of the annotation data;

the target field is used for being called when the annotation interface is rendered so as to render the annotation data into the annotation interface based on the annotation parameters.

In this manner, the client (web front end) may obtain the annotation data in response to the annotation operation of the first user on the target data, and the annotation parameters of the annotation data on one of the annotation data, where the annotation parameters are mainly used to describe the annotation data, so that the annotation data can be restored according to the annotation parameters. The client may transmit the labeling parameters to the web server, and the web server may store the labeling parameters in a target field of the target data, where the target field may be a field that describes a summary of description information of the target data, and if the target data is an image, the target field may be 1-15 fields in the description information of the image, where the labeling parameters may be stored in the target field, and specifically, the labeling parameters may be written in the target field.

When the opening operation of the target data is responded again, as the second state of the target data is set to be the marked state, the second state of the target data is indicated to have marked data, and in the case that the marked data needs to be displayed on the marked interface, one mode is that the web server can send the target data and the marked interface to the client, and the client can render the marked data into the marked interface according to the content in the target field, such as the position coordinates and the labels of the marked frame. The web server firstly renders the annotation data into the annotation interface according to the content in the target field, forms script data of the annotation interface after setting the editing state of each display content, transmits the script data to the client, and renders the corresponding annotation interface according to the script data at the client.

Based on the same inventive concept, the present disclosure further provides a display processing device for labeling an interface, with reference to fig. 9, a schematic structural diagram of the device is shown, and as shown in fig. 9, the device includes the following modules:

Optionally, the status information acquisition module includes:

an interface calling unit, configured to trigger a call of a first interface and a call of a second interface in response to the opening operation, where the first interface is configured to obtain the first state, and the second interface is configured to obtain the second state;

And the information determining unit is used for determining the state information based on the states acquired by the first interface and the second interface respectively.

Optionally, the information determining module is specifically configured to:

Optionally, the apparatus further comprises:

the state updating module is used for setting a first state of the target data as a state being marked in the process of marking the target data in the marking interface by the first user;

The interface pushing module is used for responding to the opening operation of a second user on the target data and pushing an annotation interface corresponding to the second user based on the state being annotated;

Optionally, the apparatus further comprises:

The annotation data acquisition module is used for responding to the annotation operation of the first user on the target data in the annotation interface under the condition that the editing states corresponding to the display contents are editable states, and acquiring annotation data corresponding to the annotation operation; wherein the annotation data comprises text data and/or graphic data;

The marking parameter acquisition module is used for acquiring marking parameters corresponding to the marking data; the annotation parameters are used for describing the annotation data;

the annotation file acquisition module is used for generating an annotation file corresponding to the target data based on the annotation parameters and the identification of the target data;

And the annotation file sending module is used for packaging the annotation file according to a preset format, associating the packaged annotation file with the target data and uploading the annotation file to the file storage server.

Optionally, the apparatus further comprises:

the labeling parameter acquisition module is used for responding to the labeling operation of the first user on the target data and acquiring corresponding labeling parameters based on the labeling data for labeling the target data;

The storage module is used for storing the labeling parameters into target fields of the target data; the target field is used for describing the position, the attribute and the identification of the annotation data;

Optionally, the apparatus further comprises:

The right acquisition module is used for acquiring the operation right of the first user;

the information determining module is specifically configured to determine, based on the operation right and the state information, a plurality of display contents to be displayed, and edit states corresponding to the plurality of display contents;

Optionally, the apparatus further comprises:

the subtask determining module is used for determining a target labeling subtask corresponding to the first user in a plurality of labeling subtasks;

The information determining module is specifically configured to determine, based on the state information and the target labeling subtasks, multiple display contents to be displayed, and editing states of the multiple display contents corresponding to each other in a labeling interface of the target data;

Optionally, the apparatus further comprises:

The annotation type determining module is used for determining the annotation type corresponding to the target data;

The information determining module is specifically configured to determine, based on the annotation type and the state information, a plurality of display contents to be displayed, and edit states corresponding to the plurality of display contents;

Optionally, the apparatus further comprises:

The first processing module is used for outputting the labeling interface to the first user and the fifth user according to the editing states corresponding to the display contents when the first user executes the opening operation on the target data and the first state of the target data is determined to be the state which is not being labeled when the first user detects that the fifth user executes the opening operation on the target data;

The second processing module is used for setting the editing state of the display content in the labeling interface corresponding to the target user to be an editable state for the target user which performs labeling operation in advance in the first user and the fifth user so as to allow the target user to label the target data;

And the third processing module is used for setting the editing state of the display content in the corresponding labeling interface to be an uneditable state for the user who executes the labeling operation later in the first user and the fifth user so as to prohibit the user who executes the labeling operation later from labeling the target data.

Optionally, the apparatus further comprises:

The fourth processing module is configured to, when it is detected that the first user performs the opening operation on the target data at the same time when the first user performs the opening operation on the target data, output the labeling interface to the first user and the fifth user according to editing states corresponding to the display contents if it is determined that the first state of the target data is not being labeled;

The fifth processing module is used for obtaining labeling data of labeling target users which finish labeling operation in the first user and the fifth user;

And the sixth processing module is used for generating a labeling file corresponding to the target data based on the labeling data.

The embodiment of the disclosure also provides an electronic device, and a stored computer program causes a processor to execute the display processing method of the labeling interface.

The embodiment of the disclosure also provides a computer readable storage medium, and a stored computer program causes a processor to execute the display processing method of the labeling interface according to the embodiment of the disclosure.

In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.

It will be apparent to those skilled in the art that embodiments of the present disclosure may be provided as a method, apparatus, or computer program product. Accordingly, the disclosed embodiments may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present disclosure may take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

Embodiments of the present disclosure are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the disclosure. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

While the preferred embodiments of the disclosed embodiments have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the disclosed embodiments.

Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or terminal device that comprises the element.

The foregoing has described in detail the method, apparatus, device and storage medium for display processing of a labeling interface provided by the present disclosure, and specific examples have been applied to illustrate the principles and embodiments of the present disclosure, and the description of the foregoing examples is only for aiding in understanding the method and core idea of the present disclosure; meanwhile, as one of ordinary skill in the art will have variations in the detailed description and the application scope in light of the ideas of the present disclosure, the present disclosure should not be construed as being limited to the above description.

Claims

1. A display processing method of a labeling interface, the method comprising:

and displaying a plurality of display contents according to the editing state in an annotation interface pushed to the first user.

2. The display processing method according to claim 1, wherein the acquiring the state information of the target data in response to the opening operation of the target data by the first user includes:

3. The display processing method according to claim 1, wherein the determining, based on the status information, a plurality of display contents to be displayed, and editing statuses corresponding to the plurality of display contents, respectively, includes:

4. The display processing method according to claim 1, wherein in the case where the second state indicates that the target data has been annotated, annotation data of the target data is further included in a plurality of display contents; wherein, in the case that the first state characterizes that the target data is not being marked, the editing state of the marking data is an editable state.

5. The display processing method according to claim 1, wherein in a case where an edit status of a plurality of display contents in the annotation interface is an editable and usable status, the method comprises:

6. The display processing method according to claim 1, wherein after the plurality of display contents are displayed in the editing state in the annotation interface, the method further comprises:

7. The display processing method according to claim 1, wherein after the plurality of display contents are displayed in the editing state in the annotation interface, the method further comprises:

8. The display processing method according to claim 1, characterized in that the method further comprises:

Acquiring the operation authority of the first user;

9. The display processing method according to claim 1, wherein the labeling task corresponding to the target data includes a plurality of labeling subtasks; the method further comprises the steps of:

And displaying the display content of which the corresponding editing state is in an editable state in the labeling interface as the display content under the target labeling subtask.

10. The display processing method according to claim 1, characterized in that the method further comprises:

determining the annotation type corresponding to the target data;

11. The display processing method according to claim 1, wherein when it is detected that a fifth user performs an open operation on the target data at the same timing as the first user performs the open operation on the target data, the method further comprises:

12. The display processing method according to claim 1, wherein when it is detected that a fifth user performs an open operation on the target data at the same timing as the first user performs the open operation on the target data, the method further comprises:

13. A display processing device for a labeling interface, the device comprising:

14. An electronic device, comprising: memory, a processor and a computer program stored on the memory and executable on the processor, which when executed implements a method of display processing of an annotation interface according to any of claims 1-12.

15. A computer-readable storage medium storing a computer program for causing a processor to execute the display processing method of the annotation interface according to any one of claims 1 to 12.