CN116310254A - Point cloud image labeling method, system and storage medium - Google Patents
Point cloud image labeling method, system and storage medium Download PDFInfo
- Publication number
- CN116310254A CN116310254A CN202310187424.2A CN202310187424A CN116310254A CN 116310254 A CN116310254 A CN 116310254A CN 202310187424 A CN202310187424 A CN 202310187424A CN 116310254 A CN116310254 A CN 116310254A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- data
- cloud image
- annotation
- labeling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000002372 labelling Methods 0.000 title claims abstract description 71
- 238000003860 storage Methods 0.000 title claims abstract description 16
- 238000000034 method Methods 0.000 claims abstract description 47
- 230000000694 effects Effects 0.000 claims abstract description 13
- 238000007781 pre-processing Methods 0.000 claims abstract description 11
- 238000007689 inspection Methods 0.000 claims description 16
- 238000001514 detection method Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 5
- 238000004040 coloring Methods 0.000 claims description 2
- 238000005096 rolling process Methods 0.000 claims description 2
- 238000013341 scale-up Methods 0.000 claims description 2
- 238000012545 processing Methods 0.000 abstract description 6
- 230000006870 function Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 230000004048 modification Effects 0.000 description 6
- 238000012986 modification Methods 0.000 description 6
- 230000008439 repair process Effects 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2219/00—Indexing scheme for manipulating 3D models or images for computer graphics
- G06T2219/20—Indexing scheme for editing of 3D models
- G06T2219/2012—Colour editing, changing, or manipulating; Use of colour codes
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention relates to the technical field of image data processing, in particular to a point cloud image annotation method, a point cloud image annotation system and a storage medium, wherein the method comprises the following steps: transmitting the annotation data into the task stream based on a preset requirement; configuring a label corresponding to the task and initializing a labeling authority; preprocessing the input target point cloud image and outputting a 3D color frame; and generating a point cloud image with the labeling effect display by the labeled 3D color frame. The method can realize high-efficiency and accurate labeling of a large number of 3D point cloud images of automatic driving, can greatly reduce the labeling time, and can also meet the diversified customization demands of operators and managers.
Description
Technical Field
The invention belongs to the technical field of image data processing, and particularly relates to a point cloud image annotation method, a point cloud image annotation system and a storage medium.
Background
In order to promote development and progress of automatic driving technology, each automobile industry relies on a visual model to improve target detection precision and accuracy of an automobile in a road driving process. At present, the industry has started a large amount of 2D,3D image labeling work to help to improve the model effect. The demand for the 3D point cloud labeling data is large, so that the demand for the standardized labeling work of the 3D point cloud image is provided. The patent disclosed in the prior art has less description on point cloud image labeling, and mainly focuses on a data processing direction, for example, a point cloud labeling system and a point cloud labeling method for ground identification provided by application number 202210274006.2. The patent label is single in type, the form is more suitable for semantic segmentation type labels, and the meaning of the guidance on the 3D target frame drawing is weaker. And in the background of needing mass point cloud data annotation, how to display an annotation system and an annotation method which are more suitable for 3D target frame drawing from the angle of improving efficiency is lacking.
Disclosure of Invention
The purpose of the invention is that: aiming at providing a point cloud image labeling method, a point cloud image labeling system and a storage medium, which are used for solving the problems of complex target detection labeling, easy error and low calibration efficiency of a large amount of point cloud images required by the existing automatic driving technology
In order to achieve the technical purpose, the invention adopts the following technical scheme:
in a first aspect, the present application provides a point cloud image labeling method, which is characterized by comprising the following steps:
s1, transmitting annotation data into a task stream based on preset requirements;
s2, configuring a point cloud label corresponding to the task and initializing a labeling authority;
s3, preprocessing the input target point cloud image and outputting a 3D color frame;
and S4, generating the marked 3D color frame into a point cloud image with marking effect display.
With reference to the first aspect, in some optional embodiments, the labeling data includes point cloud data and picture data, and the preset requirement includes uniformly saving formats of the point cloud data and the picture data to folders and ignoring extensions of the point cloud data and the picture data, respectively.
With reference to the first aspect, in some optional embodiments, the output format of the point cloud label is named in a character string format and is consistent with the names of the matched point cloud data, and configuring the point cloud label includes: adding or deleting the category of the point cloud label, configuring the color of the point cloud label and adjusting the display sequence of the point cloud label based on the labeling requirement of the task.
With reference to the first aspect, in some optional embodiments, the preprocessing includes: the method comprises the steps of point cloud and picture display, picture scaling, picture scrolling and dragging, picture and point cloud switching and region selection.
With reference to the first aspect, in some optional embodiments, the displaying of the point cloud and the picture includes loading the point cloud data and the picture data and displaying the matched target point cloud image and the matched target point cloud picture after initializing the labeling authority; the picture scaling is to scale up or scale down a target picture; the rolling and dragging of the picture is to move the target point cloud image; the picture and the point cloud switching is to perform the same-screen switching on the target point cloud image and the target picture; the region selection comprises the following steps: based on the labeling requirement of the task, selecting areas of the cloud image of the target point, and coloring and distinguishing different areas by different 3D color frames; the three-view presentation includes: and displaying the selected region of the target point cloud image based on the dimension of the three views.
With reference to the first aspect, in some optional embodiments, the method further includes: and configuring the state of single annotation data, configuring a locking state, a state to be checked and a checked state for the acquired annotation data in the task based on the flow, and configuring a checked state for the submitted annotation data.
With reference to the first aspect, in some optional embodiments, the point cloud image with a labeling effect is used to assist in adjusting and checking a position of a labeling result, and assist in outputting a position and a center point coordinate of a target color frame that can be mapped with the picture
With reference to the first aspect, in some optional embodiments, the method further includes checking, auditing, and quality checking the annotated data.
With reference to the first aspect, in some optional embodiments, the checking consists in: checking and modifying the problems of missing marks, multiple marks and false marks in the marked data; the auditing includes: returning the annotation data with annotation marks based on the form of graphic annotations or text annotations; the quality inspection comprises the following steps: and checking the marked data which has finished checking again and returning the marked data which still has problems based on a preset proportion.
In a second aspect, the present application provides a point cloud image annotation system comprising:
the importing module is used for transmitting the annotation data into the task stream based on the preset requirement;
the task configuration module configures a label corresponding to a task and initializes labeling authority;
the region operation module is used for preprocessing the input target point cloud image and outputting a 3D color frame;
and the detection module is used for generating the marked 3D color frame into a point cloud image with marking effect display.
And the labeling function module is used for auditing the labeling data and checking the quality of the labeling data which completes the auditing.
In a third aspect, the present application provides a computer storage medium, wherein the computer storage medium has stored therein a computer program, which when run on a computer is capable of performing the point cloud image annotation method as described above.
The invention adopting the technical scheme has the following advantages:
the method can realize high-efficiency and accurate labeling of a large number of 3D point cloud images of automatic driving, can greatly reduce the labeling time, and can also meet the diversified customization demands of operators and managers. According to the method, the complexity of the 3D point cloud data annotation scene is considered, when a plurality of persons annotate, the annotation strategy is difficult to unify, and the situation of mislabeling and missing of the annotation is serious, the annotation function is also configured, and the data with annotation marks are returned to the corresponding operators through the graph or text annotation mode, so that repair and submission are carried out. In addition, in order to improve the labeling quality of the 3D point cloud data, the invention additionally increases a round of quality inspection, and a data manager performs final inspection before checking and accepting the data which pass the checking through a certain proportion, so that various requirements of the manager are met.
Drawings
The invention can be further illustrated by means of non-limiting examples given in the accompanying drawings;
FIG. 1 is a schematic flow chart of a method for annotating a point cloud image in an embodiment of the present application;
FIG. 2 is a schematic diagram of a file format in a method for annotating a point cloud image according to an embodiment of the present application;
fig. 3 is a schematic diagram of a point cloud image labeling device in an embodiment of the present application;
fig. 4 is a schematic diagram of a point cloud image labeling device in an embodiment of the present application.
The main reference numerals are as follows:
10: an import module; 20: a task configuration module; 30: a zone operation module; 40: a detection module; 50: and marking the functional module.
Detailed Description
The present application will be described in detail below with reference to the drawings and the specific embodiments, and it should be noted that in the drawings or the description of the specification, similar or identical parts use the same reference numerals, and implementations not shown or described in the drawings are in a form known to those of ordinary skill in the art. In the description of the present application, the terms "first," "second," and the like are used merely to distinguish between descriptions and are not to be construed as indicating or implying relative importance.
Referring to fig. 1, the application provides a point cloud image annotation method, which comprises the following steps:
s1, transmitting annotation data into a task stream based on preset requirements;
s2, configuring a point cloud label corresponding to the task and initializing a labeling authority;
s3, preprocessing the input target point cloud image and outputting a 3D color frame;
and S4, generating the marked 3D color frame into a point cloud image with marking effect display.
In step S1, annotation data is transmitted into the task stream based on a preset requirement. And transmitting the required labeling data set into the task stream according to the preset requirement. The annotation data includes point cloud data and picture data. The preset requirements comprise unified data formats, and the point cloud image names and the corresponding picture names are kept consistent. The labeling system has extremely strict requirements on the format of the 3D point cloud image, and the data needs to be preprocessed before the data is uploaded into the task flow. Firstly, two folders need to be prepared, one folder stores point cloud data in a pcd format, the other folder stores picture data in a pic format, the data in the two folders are in a group every two, extension names are ignored, and naming consistency is ensured.
In step S2, a label corresponding to the task is configured and the labeling authority is initialized. That is, the labels required for the task are configured and the operator with labeling authority is initialized. After the data is prepared according to the above requirements, the data can be imported through the data transmission path. The output format of the point cloud tag file is txt type, named in six-bit character string format, is the same as the name of the corresponding point cloud file, and the data type is floating point type. The data set folder structure is shown in fig. 2.
In step S3, the entered target point cloud image is preprocessed and a 3D color frame is output. The preprocessing consists in carrying out the following operation processing on the recorded target 3D point cloud image:
a. and (3) displaying a point cloud and a picture: the task configuration module is initialized, point cloud data and picture data are loaded, and matched point cloud images and pictures are correctly displayed to an operator;
b. picture scaling: the operation of enlarging and reducing the target picture is suitable for the condition of overlarge or overlarge, and for the local characteristics, the correct display of each group of point clouds is realized, so that an operator can clearly know the complete position of each labeling object on the point cloud image, and the accurate region frame selection is performed;
c. scrolling and dragging of pictures: providing operation of moving a cloud image of a target point, and performing 360-degree movement operation through a scroll bar or a mouse, so as to perform region frame selection of accuracy rate on data at a far position;
d. picture and point cloud switching: the on-screen switching of the cloud image and the picture of the target point is realized, and a reference is provided for an operator, so that the operator can quickly and conveniently select an accurate region frame by means of the 2D picture;
e. region selection: the operator performs region selection on the target point cloud image according to the labeling task requirement, and different target regions are colored and distinguished by different 3D color frames;
f. three-view display: the selected 3D point cloud area is intuitively displayed in 3 dimensions from a front view, a side view and a top view, so that an operator can conveniently adjust and modify the 3D point cloud area, and the labeling efficiency is improved;
in step S4, the marked 3D color frame is generated into a point cloud image with a marking effect display. The marked 3D target frame is generated into a point cloud image with marking effect display, through the newly generated point cloud image, an operator can intuitively see the position distribution and the corresponding classification of each region after marking, and meanwhile, the position of the marking result is adjusted and checked, so that the accuracy of the standard result is ensured, the position, the center point and the vertex coordinates of the target frame are helped to be output, and when the data are used, the mapping can be realized through the parameters and the 2D image.
When the method is actually used, when an operator enters the labeling interface, the front view of the point cloud image is placed in the middle of the page in the reverse direction by default in the main image display area, all frame types of the 3D target detection frame are displayed above the page, the operator labels the image by using a toolbar above the page, when the operator selects a certain labeling type, the arrow is converted into "+", the arrow is dragged to the right, a 3D frame is automatically generated, and at the moment, the operator needs to click to switch or finish drawing the frame according to C. If the type of the marking frame needs to be modified, clicking the corresponding frame number on the right side of the page to modify. Meanwhile, the corresponding frame on the point cloud image can be highlighted, an operator can adjust the display view angle of the point cloud by using the left key of the mouse, the placement position of the point cloud image is positioned by dragging the right key, the size and the position of the point cloud image are scaled by using the mouse wheel to view more details, and the operations of checking, modifying and the like on the labeling details are facilitated.
And under the page, whether the three views of the 3D point cloud frame are completely selected or not is observed, the size and the direction of the 3D frame are adjusted, the direction of the 3D frame is ensured to be consistent with the range of the target point cloud, and the target vehicle or an object to be marked is ensured to be consistent with the actual direction through direction adjustment.
The method further comprises the steps of: marking information statistics, marking auditing and quality inspection.
And (5) marking information statistics. In the invention, when the task is configured and the right is initialized, the label category can be added or deleted according to the label requirement of the current task, and the customization modification is carried out on different labels, including configuration colors, display sequence adjustment, and in addition, the functional design of the added options is more beneficial to personalized presentation of the labels, thereby meeting the multi-aspect requirement of operators.
The labeling system is provided with the function of checking the task state and the single data state, so that the labeling work progress can be checked conveniently. By displaying the remaining number of overall tasks, the lifecycle of the current project is monitored for health. And locking, checking and quality-checked data acquired or submitted in the task.
And marking the page, namely, displaying the length, the width and the height of the 3D frame, namely, the three-dimensional coordinates of the center point of the 3D marking frame, and after marking, deriving all information for algorithm use. The system has the function of systematically managing the image information of the whole system.
Checking, auditing and quality inspection. The modification label is set for the labeled data, and errors such as label error, label missing, multiple labels or error labels occur due to fatigue or fuzzy rules during labeling, so that the labeled data set needs to be checked and modified. The module provides a corresponding modification function, and after all the labeling data are labeled, if the wrong labeling is found, the real-time modification can be carried out, and the modified labeling data are submitted. In addition, the method and the system consider the complexity of the 3D point cloud data annotation scene, and when the annotation strategy is difficult to unify and the error mark and omission mark are serious in the multi-person annotation process, an annotation function is also configured, and the data with annotation marks are returned to corresponding operators through the graph or text annotation mode for repair and submission. In addition, in order to improve the labeling quality of the 3D point cloud data, a round of quality inspection is additionally added, a data manager performs final inspection before acceptance of the approved labeling data according to a certain proportion, and the data with problems after the inspection can be modified and stored by 4 ways of returning to the original labeling person, returning to the original inspector, returning to an inspection pool, returning to labels and the like, so that various requirements of the manager are met.
Referring to fig. 3 and 4, an embodiment of the present application provides a point cloud image annotation System, which includes at least one software function module stored in a memory module in the form of software or Firmware (Firmware) or solidified in an Operating System (OS) in the annotation System. Each module is used for executing executable modules stored in the storage module, such as a software functional module, a computer program module and the like included in the point cloud image annotation system.
The system comprises an application module and a labeling module, wherein the application module comprises an importing module 10, a task configuration module 20, a region operation module 30 and a detection module 40. The labeling module includes a labeling function module 50. The functions of each module are as follows:
the importing module 10 transmits the annotation data into the task stream based on a preset requirement;
the task configuration module 20 configures a label corresponding to a task and initializes labeling authority;
the region operation module 30 is used for preprocessing the input target point cloud image and outputting a 3D color frame;
the detection module 40 generates a point cloud image with the labeling effect display from the labeled 3D color frame.
The labeling function module 50 performs inspection, auditing and quality inspection on the labeled data.
In this embodiment, the memory module may be, but is not limited to, a random access memory, a read-only memory, a programmable read-only memory, an erasable programmable read-only memory, an electrically erasable programmable read-only memory, etc. In this embodiment, the storage module may be used to store the annotation data in the import module. Of course, the storage module may also be used to store a program, and the processing module executes the program after receiving the execution instruction.
It will be appreciated that the point cloud image annotation system shown in figure 4 is merely a schematic structural representation and that the point cloud image annotation system may comprise more components than those shown in figure 4. The components shown in fig. 4 may be implemented in hardware, software, or a combination thereof.
It should be noted that, for convenience and brevity of description, the above-described importing process, configuration process, operation process, detection process, etc. may refer to the corresponding process of each step in the foregoing method, and will not be described in detail herein.
Embodiments of the present application also provide a computer-readable storage medium. The computer readable storage medium has stored therein a computer program which, when run on a computer, causes the computer to perform the point cloud image annotation method as described in the above embodiments.
From the foregoing description of the embodiments, it will be apparent to those skilled in the art that the present application may be implemented in hardware, or by means of software plus a necessary general hardware platform, and based on this understanding, the technical solution of the present application may be embodied in the form of a software product, where the software product may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disc, a mobile hard disk, etc.), and includes several instructions to cause a computer device (may be a personal computer, an emulation device, or a network device, etc.) to perform the methods described in the respective implementation scenarios of the present application.
In summary, the present application provides a method, a system and a storage medium for annotating point cloud images, where the method includes the following steps: transmitting the annotation data into the task stream based on a preset requirement; configuring a label corresponding to the task and initializing a labeling authority; processing the input target point cloud image based on preprocessing; and generating a point cloud image with the labeling effect display by the labeled target frame. The method can realize high-efficiency and accurate labeling of a large number of 3D point cloud images of automatic driving, can greatly reduce the labeling time, and can also meet the diversified customization demands of operators and managers. According to the method, the complexity of the 3D point cloud data annotation scene is considered, when a plurality of persons annotate, the annotation strategy is difficult to unify, and the situation of mislabeling and missing of the annotation is serious, the annotation function is also configured, and the data with annotation marks are returned to the corresponding operators through the graph or text annotation mode, so that repair and submission are carried out. In addition, in order to improve the labeling quality of the 3D point cloud data, the invention additionally increases a round of quality inspection, and a data manager performs final inspection before checking and accepting the data which pass the checking through a certain proportion, so that various requirements of the manager are met.
In the embodiments provided in this application, it should be understood that the disclosed system, system and method may be implemented in other manners as well. The systems, and method embodiments described above are illustrative only, and the flowcharts and block diagrams in the figures, for example, illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions. In addition, the functional modules in the embodiments of the present application may be integrated together to form a single part, or each module may exist alone, or two or more modules may be integrated to form a single part.
The method, the system and the storage medium for annotating the point cloud image provided by the invention are described in detail. The description of the specific embodiments is only intended to aid in understanding the method of the present invention and its core ideas. It should be noted that it will be apparent to those skilled in the art that various modifications and adaptations of the invention can be made without departing from the principles of the invention and these modifications and adaptations are intended to be within the scope of the invention as defined in the following claims.
Claims (11)
1. The point cloud image marking method is characterized by comprising the following steps of:
s1, transmitting annotation data into a task stream based on preset requirements;
s2, configuring a point cloud label corresponding to the task and initializing a labeling authority;
s3, preprocessing the input target point cloud image and outputting a 3D color frame;
and S4, generating the marked 3D color frame into a point cloud image with marking effect display.
2. The point cloud image annotation method according to claim 1, wherein the annotation data comprises point cloud data and picture data, and the preset requirement comprises uniformly storing formats of the point cloud data and the picture data in a folder respectively and ignoring extension names of the point cloud data and the picture data.
3. The point cloud image annotation method of claim 2, wherein the output format of the point cloud label is named in a character string format and is consistent with the names of the matched point cloud data, respectively, and configuring the point cloud label comprises: adding or deleting the category of the point cloud label, configuring the color of the point cloud label and adjusting the display sequence of the point cloud label based on the labeling requirement of the task.
4. The point cloud image annotation method of claim 1, wherein the preprocessing comprises: the method comprises the steps of point cloud and picture display, picture scaling, picture scrolling and dragging, picture and point cloud switching, region selection and three-view display.
5. The point cloud image annotation method of claim 4, wherein the point cloud and picture display is that after initializing the annotation authority, loading the point cloud data and the picture data and displaying matched target point cloud images and pictures; the picture scaling is to scale up or scale down a target picture; the rolling and dragging of the picture is to move the target point cloud image; the picture and the point cloud switching is to perform the same-screen switching on the target point cloud image and the target picture; the region selection comprises the following steps: based on the labeling requirement of the task, selecting areas of the cloud image of the target point, and coloring and distinguishing different areas by different 3D color frames; the three-view presentation includes: and displaying the selected region of the target point cloud image based on the dimension of the three views.
6. The point cloud image annotation method of claim 1, further comprising: and configuring the state of single annotation data, configuring a locking state, a state to be checked and a checked state for the acquired annotation data in the task based on the flow, and configuring a checked state for the submitted annotation data.
7. The method of claim 6, wherein the point cloud image with labeling effect is used for assisting in adjusting and checking the position of the labeling result and assisting in outputting the position and the center point coordinates of the target color frame capable of being mapped with the picture.
8. The point cloud image annotation method of claim 7, further comprising checking, auditing and quality inspection of the annotated data.
9. The point cloud image annotation method of claim 8, wherein the checking consists in: checking and modifying the problems of missing marks, multiple marks and false marks in the marked data; the auditing includes: returning the annotation data with annotation marks based on the form of graphic annotations or text annotations; the quality inspection comprises the following steps: and checking the marked data which has finished checking again and returning the marked data which still has problems based on a preset proportion.
10. A point cloud image annotation system comprising:
the importing module (10) is used for transmitting the annotation data into the task stream based on preset requirements;
the task configuration module (20) configures a label corresponding to the task and initializes the labeling authority;
the region operation module (30) is used for preprocessing the input target point cloud image and outputting a 3D color frame;
the detection module (40) generates a point cloud image with the labeling effect display from the labeled 3D color frame;
and the labeling function module (50) is used for checking, auditing and quality inspection of the labeled data.
11. A computer storage medium having a computer program stored therein, which, when run on a computer, is capable of performing the point cloud image annotation method according to any of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310187424.2A CN116310254A (en) | 2023-03-02 | 2023-03-02 | Point cloud image labeling method, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310187424.2A CN116310254A (en) | 2023-03-02 | 2023-03-02 | Point cloud image labeling method, system and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116310254A true CN116310254A (en) | 2023-06-23 |
Family
ID=86835444
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310187424.2A Pending CN116310254A (en) | 2023-03-02 | 2023-03-02 | Point cloud image labeling method, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116310254A (en) |
-
2023
- 2023-03-02 CN CN202310187424.2A patent/CN116310254A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10157502B2 (en) | Method and apparatus for sharing augmented reality applications to multiple clients | |
| CN102156769B (en) | Multi-dimensional artifact assemblage for infrastructure and other assets with interface node mediators | |
| Karsch et al. | ConstructAide: analyzing and visualizing construction sites through photographs and building models | |
| Gimeno et al. | A new AR authoring tool using depth maps for industrial procedures | |
| CN110414307A (en) | The method and system divided for semi-automatic image and annotated | |
| Zollmann et al. | Interactive 4D overview and detail visualization in augmented reality | |
| US10997782B2 (en) | Associating a point cloud image with a piping line drawing and applying dimensions regarding the same | |
| CN110503581B (en) | Unity 3D-based visual training system | |
| CN105159869A (en) | Picture editing method and system | |
| CN115424265A (en) | Point cloud semantic segmentation and labeling method and system | |
| CN114565908A (en) | Lane line detection method and device, electronic device and storage medium | |
| CN114612393A (en) | Monocular vision-based reflective part pose estimation method | |
| US11488332B1 (en) | Intensity data visualization | |
| EP3282351A1 (en) | System and method for facilitating an inspection process | |
| CN112346807A (en) | Image annotation method and device | |
| Köppel et al. | Context-responsive labeling in augmented reality | |
| CN112799583A (en) | Point location schematic drawing system based on android platform | |
| CN109887054B (en) | Method and device for displaying map | |
| CN116310254A (en) | Point cloud image labeling method, system and storage medium | |
| Kikuchi et al. | Diminished reality using semantic segmentation and generative adversarial network for landscape assessment: evaluation of image inpainting according to colour vision | |
| CN111272186A (en) | Robot path configuration method and device and readable storage medium | |
| CN116245052A (en) | Drawing migration method, device, equipment and storage medium | |
| CN116129126A (en) | Construction method of segmentation prediction model, picture semantic segmentation labeling method and device | |
| CN115857930A (en) | Prototype rapid design method and system based on real-scene model | |
| Gimeno et al. | An easy-to-use AR authoring tool for industrial applications |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |