CN110147701B - Key point labeling method, device, computer equipment and storage medium - Google Patents
Key point labeling method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN110147701B CN110147701B CN201810678758.9A CN201810678758A CN110147701B CN 110147701 B CN110147701 B CN 110147701B CN 201810678758 A CN201810678758 A CN 201810678758A CN 110147701 B CN110147701 B CN 110147701B
- Authority
- CN
- China
- Prior art keywords
- labeling
- key point
- picture
- sample picture
- instruction
- 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.)
- Active
Links
- 238000002372 labelling Methods 0.000 title claims abstract description 205
- 238000000034 method Methods 0.000 claims abstract description 92
- 230000008569 process Effects 0.000 claims abstract description 61
- 230000006870 function Effects 0.000 claims description 41
- 238000012545 processing Methods 0.000 claims description 13
- 230000015654 memory Effects 0.000 claims description 6
- 239000003086 colorant Substances 0.000 claims description 4
- 238000007667 floating Methods 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 238000012549 training Methods 0.000 abstract description 13
- 230000003993 interaction Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 13
- 238000012544 monitoring process Methods 0.000 description 7
- 230000009471 action Effects 0.000 description 6
- 210000003811 finger Anatomy 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 230000033001 locomotion Effects 0.000 description 5
- 210000001503 joint Anatomy 0.000 description 3
- 230000009191 jumping Effects 0.000 description 3
- 210000003813 thumb Anatomy 0.000 description 3
- 238000012790 confirmation Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 210000000811 metacarpophalangeal joint Anatomy 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000001960 triggered effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 210000000707 wrist Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/40—Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/28—Recognition of hand or arm movements, e.g. recognition of deaf sign language
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Multimedia (AREA)
- Collating Specific Patterns (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The invention discloses a key point labeling method, a key point labeling device, computer equipment and a storage medium, and belongs to the technical field of man-machine interaction. According to the method, in the labeling process, the sample picture to be labeled is displayed, and meanwhile, the display of the labeling reference picture is provided, so that a labeling person can guide the labeling process based on the position, the number information and the like of each key point in the labeling reference picture, the labeling person is assisted to judge whether the labeling is correct or not, the labeling accuracy and the labeling authenticity can be effectively improved, and the recognition accuracy of a training model is improved.
Description
Technical Field
The present invention relates to the field of man-machine interaction technologies, and in particular, to a method and apparatus for labeling key points, a computer device, and a storage medium.
Background
Gesture recognition is an important technique in human-computer interaction by which a user can trigger a computer device to perform certain functions. In order to recognize the gesture operation of the user, a deep learning model for recognizing the gesture needs to be trained, and a large amount of training data is needed during training, wherein the training data can be a picture marked with key points of the gesture, so how to mark the key points of the picture is important.
At present, when the key points are marked, the motion simulation engine knowledge and related programming technology are mainly applied to simulate the motions of different people in different scenes, the positions of the key points of the gesture of the people in different scenes can be directly obtained, and then after the key points are marked, a plurality of key points can be connected to obtain key point lines, and the gesture skeleton is represented by the lines.
When a person is simulated by using a computer technology, the simulated gesture of the person has a certain difference from the actual gesture, but the reality and the accuracy of the gesture key points marked based on the simulation are poor, and the recognition accuracy of the trained model is low.
Disclosure of Invention
The embodiment of the invention provides a key point labeling method, a device, computer equipment and a storage medium, which can solve the problems of authenticity and accuracy of a related technology in the process of key point labeling, thereby improving the recognition accuracy of a training model. The technical scheme is as follows:
In one aspect, a method for labeling key points is provided, the method comprising:
Displaying a labeling reference picture and a sample picture, wherein the labeling reference picture comprises a plurality of first key points and the number information of each first key point, and the number information of each first key point is used for indicating the labeling sequence for labeling the key points;
Marking a plurality of second key points on the sample picture according to a plurality of triggering operations on the sample picture;
And acquiring a labeling result file of the sample picture based on the labeled second key points in the sample picture.
In one aspect, a key point labeling device is provided, the device includes:
the display module is used for displaying a labeling reference picture and a sample picture, wherein the labeling reference picture comprises a plurality of first key points and the number information of each first key point, and the number information of each first key point is used for indicating the labeling sequence for labeling the key points;
The labeling module is used for labeling a plurality of second key points on the sample picture according to a plurality of triggering operations on the sample picture;
And the storage module is used for acquiring a labeling result file of the sample picture based on the labeled second key points in the sample picture.
In one aspect, a computer device is provided that includes a processor and a memory having at least one instruction stored therein that is loaded and executed by the processor to perform operations as performed by the keypoint labeling method described above.
In one aspect, a computer-readable storage medium having stored therein at least one instruction that is loaded and executed by a processor to implement operations performed by a keypoint tagging method as described above is provided.
The technical scheme provided by the embodiment of the invention has the beneficial effects that:
In the labeling process, the sample picture to be labeled is displayed and the display of the labeling reference picture is provided, so that a labeling person can guide the labeling process based on the position, the number information and the like of each key point in the labeling reference picture, the labeling person is assisted to judge whether the labeling is correct or not, the labeling accuracy and the labeling authenticity can be effectively improved, and the recognition accuracy of the training model is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is an implementation scene diagram of a key point labeling method provided by an embodiment of the present invention;
FIG. 2 is a flowchart of a method for labeling key points according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an annotation interface provided by an embodiment of the present invention;
FIG. 4 is a schematic illustration of yet another labeling interface provided by an embodiment of the present invention;
FIG. 5 is a schematic diagram of an encrypted string according to an embodiment of the present invention;
FIG. 6 is a diagram of a display interface of a marked picture according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of a process from importing an application program to confirming that a labeling is successful for a sample picture to be labeled according to an embodiment of the present invention;
FIG. 8 is a schematic structural diagram of a key point labeling device according to an embodiment of the present invention;
FIG. 9 is a schematic structural diagram of an implementation of a key point labeling device according to an embodiment of the present invention;
fig. 10 is a schematic structural diagram of a computer device according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the embodiments of the present invention will be described in further detail with reference to the accompanying drawings.
Fig. 1 is an implementation scenario diagram of a key point labeling method provided by an embodiment of the present invention. In this implementation scenario, the implementation scenario includes at least one computer device 101 and at least one server 102, where the at least one computer device 101 may include a computer device for performing a key point labeling, may also include a computer device for collecting a gesture image, and the embodiment of the present invention is not limited to this, and the at least one server 102 may include a server for training a gesture recognition model, and may also include a server for applying a gesture recognition model, and of course, the above computer device may also be configured with a trained gesture recognition model to implement a gesture recognition function. Communication between the computer device 101 and the server 102 may be via a wired or wireless network, so that the computer device 101 may send the annotated pictures to the server for model training.
Fig. 2 is a flowchart of a method for labeling key points according to an embodiment of the present invention. Referring to fig. 2, this embodiment specifically includes:
201. The method comprises the steps that a computer device displays a labeling reference picture and a sample picture, wherein the labeling reference picture comprises a plurality of first key points and the number information of each first key point, and the number information of each first key point is used for indicating the labeling sequence for labeling the key points.
The computer device can be provided with an application program for realizing the key point labeling function, and when a user triggers an operation instruction of the application program, the computer device can respond to the operation instruction to display a labeling reference picture and a sample picture to be labeled.
In one possible implementation, the application may provide a plurality of different types of keypoint labeling functions, such as a gesture keypoint labeling function, a face keypoint labeling function, a human skeletal keypoint labeling function, and any of the other picture labeling functions, and accordingly, the computer device may display an initial interface of the application in response to the run instruction, the initial interface including a plurality of function options thereon, each function option corresponding to a picture labeling function. The user can trigger any function option according to the labeling requirement of the user, and when the computer equipment detects the triggering operation of the function option, the labeling reference picture corresponding to the function option is displayed. For example, referring to fig. 3, when the computer device detects a triggering operation of a function option corresponding to a gesture key point labeling function, a labeling reference picture corresponding to the gesture labeling function is displayed.
The number information of the first key point in the marked reference picture is expressed in a numerical form, so that the numerical order can indicate an operation order, an intuitive prompting effect can be achieved, and a user only needs to mark according to the order of the number of the first key point on the marked reference picture from small to large when performing trigger operation.
When the labeling reference picture is displayed, the labeling reference picture may include a target object and a plurality of key points on the target object, and each key point is further displayed with numbering information of the key point, and still taking the gesture key point labeling function shown in fig. 3 as an example, fig. 3 shows a plurality of first key points on a human hand and a human hand, where the plurality of first key points mark a position of each key point of the hand, for example, 21 key points of a wrist, a thumb tip, a thumb first interphalangeal joint, a thumb metacarpophalangeal joint, a large thenar center, an index finger, a middle finger, a ring finger, a fingertips of small fingers, a first interphalangeal joint, a second interphalangeal joint, a metacarpophalangeal joint, and the like, and the plurality of first key points are all marked with numbering information, for example, numbers 0-20 shown in fig. 3, and a connecting line is arranged between some of the plurality of first key points to represent a preset connection relationship between the first key points. Of course, for gesture keypoint labeling, different connecting lines formed between the first keypoints of the labeling reference picture can also be displayed in different colors so as to distinguish different branches of the human hand skeleton.
In the embodiment of the invention, the sample pictures to be marked can be imported to the application program in batches by a user, the sample pictures to be marked can be imported from external storage equipment or from any storage area on the computer equipment, and when the sample pictures are imported to the application program, the sample pictures can be stored in folders corresponding to the application program. When the import is successful, the computer device may display any of the sample pictures of the batch of imported sample pictures. In the displaying process, a display manner as shown in fig. 3 may be adopted, where a labeling reference picture (such as a gesture picture in a left graph in fig. 3) is displayed on the left side of the interface, and a sample picture to be labeled is displayed on the right side of the interface, and of course, according to a usage habit of a user, a display area of the picture is exchanged, so that the labeling reference picture is displayed on the right side of the interface, and the sample picture to be labeled is displayed on the left side of the interface, that is, the method further includes: when a position exchange instruction is received, the display positions of the labeling reference picture and the sample picture are exchanged in the interface, and an interface display mode which is more in line with the use habit can be provided for a user through the exchange function. The sample picture can be a real picture of the marked object, such as a human hand picture, so that the marking authenticity is greatly improved.
Of course, in order to make the user know how to make the labeling, when the labeling reference picture and the picture to be labeled are displayed, operation prompt information can be displayed, and the operation prompt information is used for indicating how to label the picture to be labeled according to the key points displayed on the labeling reference picture so as to achieve the purpose that the user can normally use the picture. For example, the operation prompt information may be "please click on the corresponding positions in the drawing in sequence according to the numbering sequence in the left drawing".
202. When the computer equipment detects the triggering operation on the sample picture, the number information of the second key point corresponding to the triggering operation is determined according to the marked second key point.
The triggering operation may be performed based on an input device, for example, the triggering operation may be performed by a user triggering a mouse, and of course, for a touch screen, the triggering operation may also be a touch operation, that is, the user may perform the triggering operation by touching a screen with a finger.
The second key point marked refers to the second key point on the sample picture, wherein the coordinate and the number information of the second key point are determined based on the triggering operation. Determining the number information of the second key point corresponding to the triggering operation according to the marked second key point: and adding 1 to the maximum number information in the marked second key points to obtain the number information of the second key points corresponding to the triggering operation.
For the first trigger operation on the sample picture, since no label has been made on the sample picture before, the first trigger operation is labeled with the second key point with the number information of 1. For any triggering operation after the first triggering operation, the marked key point should be the next key point of the previous marked key point, that is, the number information of the newly marked key point should be one number information after the maximum number information in the marked key points, for example, the number information of the key point marked for the triggering operation before the present triggering operation is 10, and the marked number information of the present triggering operation is 11.
203. And the computer equipment determines the coordinates of the contact point of the triggering operation as the coordinates of the second key point on the sample picture, and displays the second key point and the number information of the second key point on the coordinates of the triggering point.
In the embodiment of the invention, the triggering operation corresponds to a touch point coordinate on the sample picture, the touch point coordinate refers to a click coordinate of a cursor of a mouse on the sample picture for the mouse operation, and the triggering point coordinate refers to a touch point coordinate of a finger touching on the sample picture for the touch operation. Wherein the determined coordinates may be expressed as: key points with number information 1: (X1, Y1) so that the number information and coordinates of the key points can be corresponded.
When the labeling is carried out, a user can carry out triggering operation on corresponding positions on a sample picture to be labeled according to the positions of a plurality of first key points in a labeling reference picture, the computer equipment determines the next key point of the labeled second key point as the second key point corresponding to the triggering operation according to the detected triggering operation and the labeled second key point, the coordinates of the contact point of the triggering operation are determined as the coordinates of the second key point on the sample picture, and the computer equipment displays the second key point and the number information of the second key point in the sample picture.
In the above process, each time a second key point is marked by triggering operation, the marked second key point can be displayed on the sample picture, so as to provide visual effect, and the user can check whether the mark is correct or not, so that adjustment can be performed in time.
The steps 202 to 203 are a process that the computer device marks a plurality of second key points on the sample picture according to a plurality of trigger operations on the sample picture, in this process, only how a specific key point is marked by a trigger operation is described as an example, and in practical application, the process needs to be executed for each detected trigger operation to complete marking of the key point in the sample picture.
For example, referring to FIG. 4, FIG. 4 provides a schematic illustration of an annotation interface. In the gesture key point labeling process, a user can perform trigger operation on corresponding skeleton joint positions on a sample picture to be labeled according to skeleton joint positions of a plurality of first key points in a labeling reference picture, the computer equipment determines the next key point of the labeled second key point as a second key point corresponding to the trigger operation according to the detected trigger operation and the labeled second key point, the trigger point coordinates of the trigger operation are determined as the coordinates of the second key point on the picture, and the computer equipment displays the second key point on an interface. The second key point can be overlaid on the original layer of the sample picture in a floating layer mode for display. In addition, the second key point may be represented by a color dot, such as green, etc.
204. And the computer equipment connects the second key points indicated by the preset connection relation by lines according to the preset connection relation among the second key points to obtain at least one key point line, and at least one key point connecting line is displayed on the sample picture.
For the second key points with different number information, the number information has a preset connection relationship, that is, the second key points have a preset connection relationship, and the preset connection relationship is used for showing which key points belong to the same skeleton branch. Therefore, when the labeling of the plurality of second key points of the sample picture is completed, the plurality of second key points can be connected by lines based on the preset connection relation. Of course, the connection process may be performed in the labeling process, and each time a second key point is labeled, the second key point is connected with a labeled second key point having a preset connection relationship by using a line, and is displayed on the sample picture. Optionally, the key point connecting lines on the same branch of the skeleton may be represented by the same color, and the key point connecting lines on different branches of the skeleton may be displayed by different colors, so as to distinguish the different branches of the skeleton.
It can be known from the above description that the labeling process and the displaying of the connecting line of the key points can be performed simultaneously, that is, when the user performs the triggering operation on the picture, the computer device displays the second key point corresponding to the triggering operation, and if the second key point corresponding to the triggering operation has a preset connection relationship with the labeled second key point, the computer device connects the second key point and the labeled second key point to obtain a key point line, and simultaneously displays the key point line.
205. And when a labeling completion instruction is received, the computer equipment encrypts based on the coordinates of the plurality of second key points on the picture to obtain an encrypted character string, and stores the encrypted character string into a labeling result file of the sample picture.
The computer equipment encrypts the coordinates of each second key point obtained by labeling by adopting a preset secret key and an encryption algorithm to obtain an encrypted character string so as to ensure the security of labeling data. The encryption algorithm may be a DES encryption algorithm, although other encryption algorithms may be used, which is not limited in this embodiment of the present invention.
When a user finishes marking a sample picture, the marking result of the sample picture can be stored in a marking result file. The labeling result of each picture can be stored as a labeling result file, and of course, the labeling result of the batch of pictures can also be stored into a labeling result file. Referring to fig. 5, when a user opens a labeling result file, an encrypted character string obtained by encrypting coordinates of the sample picture is displayed.
In some embodiments, when the computer device receives the labeling completion instruction, the encryption process in step 207 may not be executed, and the labeled coordinates may be directly stored as a labeling result file of the sample picture for a subsequent training process.
206. And when the computer equipment receives the picture viewing instruction, decrypting the encrypted character string in the labeling result file of the sample picture, and displaying a plurality of second key points of the sample picture, the number information of each second key point and lines obtained by connecting the plurality of second key points on the sample picture based on the data obtained after decryption.
The computer device may also provide a picture viewing function for the user to confirm the correctness of the labeling results. When a user triggers a picture viewing instruction on an application program, the computer equipment receives the picture viewing instruction and responds to the instruction to display a picture marking result in a marking result file. Referring to fig. 6, fig. 6 shows a plurality of second keypoints, numbering information, and lines connected by the plurality of second keypoints of the sample picture.
Optionally, when the coordinates are stored after being encrypted, the computer device may decrypt the coordinates with the same or matching key as step 205 based on the encrypted string in the image annotation result file, to obtain decrypted data.
It should be noted that, in the process of checking, whether the labeling is correct or not can be confirmed based on the display of the picture and the key points, if not, the labeling result file can be deleted, and the labeling operation is carried out again, so that the situation of poor labeling can be greatly reduced.
In the actual labeling process, a picture checking instruction can be triggered after each sample picture is labeled to verify whether the labeling is correct, or a picture checking instruction can be triggered again after a batch of sample pictures are labeled to verify the correctness of a sample picture which is labeled Zhang Zhanshi, and the embodiment of the invention is not limited to this.
The steps 201 to 206 describe the sequence from marking to data storage to viewing marking results for a sample picture, and in the marking process, various different processing manners may be provided, for example, deleting some or all marked key points, switching to the next picture to continue marking, exiting the marking process, and the like, and the various processing manners are described below in different cases:
(1) And deleting the marked certain key point.
For any marked key point, the user can remark the key point, that is, the previous key point is returned to be remarked, the original mark needs to be deleted, and correspondingly, the method further comprises the following steps: and deleting the number information of the previous second key point and the previous gesture key point when a first instruction is received, wherein the first instruction is used for indicating to return to the previous key point for marking again.
In the process of marking the key points, as long as the application program does not receive the exit instruction, the computer equipment can know that the user does not complete the triggering operation of all the key points in the picture, and a processing mode for modifying the marked key points can be provided. The user can trigger the first instruction of the application program through triggering operation of the first designated key, and when the computer equipment receives the first instruction, the recorded coordinates and the number information at the last triggering position are deleted. The first designated key may be a P key, and of course, may also be other keys designated in advance, which is not limited in the embodiment of the present invention.
(2) And deleting all marked key points. The computer device may also provide a quick deletion way to delete all key points by one operation, and accordingly the method further comprises: and when a second instruction is received, deleting all second key points in the sample picture and the number information of all second key points, wherein the second instruction is used for deleting all marked key points.
The user can trigger a second instruction of the application program through triggering operation of a second designated key, and when the computer equipment receives the second instruction, all the added marks in the sample picture are deleted. The second designated key may be an r key, and of course, may also be other keys designated in advance, which is not limited in the embodiment of the present invention.
(3) And switching to the next picture to continue labeling. When any sample picture is marked, the computer device can provide a continuous marking process, that is, if one marking is completed or the key point of the current picture is not complete, marking can not be realized, and the like, the current sample picture can be skipped and the marking of the next sample picture can be entered, and correspondingly, the method further comprises: and when a third instruction is received, switching and displaying the next sample picture, wherein the third instruction is used for switching the picture. And triggering a third instruction of the application program by the triggering operation of a third designated key, and switching to display the next picture to be annotated when the computer equipment receives the third instruction. The third designated key may be a k key, and of course, may also be other keys designated in advance, which is not limited in the embodiment of the present invention.
(4) And exiting the labeling process. When the user wants to end the labeling, the labeling process can be exited, and correspondingly, the method further comprises: and when a fourth instruction is received, exiting the labeling process and recording unlabeled sample pictures, wherein the fourth instruction is used for exiting the labeling process. For the exiting, the next time the application is started, the unlabeled sample picture can be directly displayed. The user can trigger a fourth instruction of the application program through triggering operation of a fourth designated key, and when the computer equipment receives the fourth instruction, the labeling process is exited and unlabeled pictures are recorded. The fourth designated key may be a q key, and of course, may also be other keys designated in advance, which is not limited in the embodiment of the present invention.
(5) The labeling of the current sample picture is completed and the labeling of the next sample picture is entered. When all key points of the sample picture are marked, marking the next sample picture can be carried out, and correspondingly, the method further comprises: and when a fifth instruction is received, saving the marked sample picture and switching to the next sample picture, wherein the fifth instruction is used for the sample picture to finish marking. The user can trigger a fifth instruction of the application program through triggering operation of a fifth designated key, and when the computer equipment receives the fifth instruction, the marked picture is saved and the next picture to be marked is switched. The fifth designated key may be an s key, and of course, may also be other keys designated in advance, which is not limited in the embodiment of the present invention.
For the process of viewing the noted sample picture, the computer device may also provide a plurality of different processing manners to adapt to the noted requirement of the user, which is specifically described as follows:
(6) And displaying the next marked picture. In the display process of any marked picture, the next picture can be switched to be displayed through switching operation, and correspondingly, the method further comprises the following steps: and when a sixth instruction is received in the display process of the sample picture, switching and displaying the next marked picture, wherein the sixth instruction is used for switching the marked picture. The user can trigger a sixth instruction of the application program through triggering operation of a sixth designated key, and when the computer equipment receives the sixth instruction, the next marked picture is switched to be displayed. The six specified keys may be n keys, and of course, may also be other keys specified in advance, which is not limited in the embodiment of the present invention.
(7) And deleting the labeling result file of the sample picture. Since the labeling of the key points may need to be modified, a deletion function of the labeling result file may also be provided, and correspondingly, the method further includes: and deleting the labeling result file of the picture when a seventh instruction is received in the display process of the sample picture, wherein the seventh instruction is used for deleting the labeling result file of the picture. The user can trigger a seventh instruction of the application program through triggering operation of a seventh designated key, and when the computer equipment receives the seventh instruction, the labeling result file of the picture is deleted. The seven designated keys may be d keys, and of course, may also be other keys designated in advance, which is not limited in the embodiment of the present invention.
It should be noted that, through the labeling process of steps 202 and 203 and the confirmation of step 206, the number of inferior labels can be greatly reduced, and the quality of the training set sample can be improved. In the actual labeling process, some of the above steps are batch-processed. For example, after any sample picture is marked successfully, in some embodiments, the method may not immediately enter a step of verifying the correctness of the marking, but marking the next sample picture, and displaying after marking all the sample pictures in a batch is finished, so as to verify the correctness.
Fig. 7is a schematic diagram of a process from importing an application program to confirming that labeling is successful, according to an embodiment of the present invention, as shown in fig. 7. Referring to fig. 7, the process may be performed based on the OpenCV vision library to create a way of interaction between a person and a computer, including:
1. And starting an application program, and displaying the marked reference picture and the sample picture.
2. Judging whether the sample picture has a labeling result file, if so, skipping the sample picture, and if not, labeling the sample picture by a user through triggering operation on the sample picture, and skipping to the step 3;
3. In the labeling process of the sample picture, if the triggering operation of the s key is detected, the labeling is finished, and the step 4 is skipped; if the triggering operation on the k key is detected, the marking is ignored, and the step 5 is skipped; pressing the q key to indicate that the program is exited, and jumping to the step 6;
4. Recording the labeling result after successful labeling, and jumping to the step 7;
5. Neglecting the labeling of the sample picture, recording the labeling result, and jumping to the step 7;
6. The program is ended, and the step1 is skipped;
7. In the display stage, if the triggering operation of the d key is detected, the marking result is required to be deleted, and the step 2 is skipped; if the triggering operation of the n key is detected, the marked picture is qualified, the next marked picture can be continuously checked, and the step 8 is skipped;
8. And determining that the picture is marked successfully.
In the process, the marking of one picture is successful, and a plurality of steps are needed, wherein the marking of the step 3 and the confirmation of the step 7 can greatly reduce the number of inferior marks and improve the quality of the training set sample.
In the actual labeling process, some of the above steps are batch-processed. For example, after the labeling of the 4 th step is successful, in some embodiments, the method may not immediately enter the 7 th step to verify the correctness of the labeling, but first label the next sample picture, and after the labeling of a batch of sample pictures is finished, display the sample pictures so as to verify the correctness.
In the labeling process, the sample picture to be labeled is displayed and the display of the labeling reference picture is provided, so that a labeling person can guide the labeling process based on the position, the number information and the like of each key point in the labeling reference picture, the labeling person is assisted to judge whether the labeling is correct or not, the labeling accuracy and the labeling authenticity can be effectively improved, and the recognition accuracy of the training model is improved.
Further, when the annotator annotates the key points, adjacent key points are connected to form a framework, so that the annotator can be assisted to judge whether the annotation is correct or not. Further, a plurality of interactive keys are provided to realize a plurality of processing functions in the labeling process and after labeling is successful, so that a annotator can tell the computer device to perform different processing, such as returning to one step, restarting or exiting the program, etc., through the interactive keys. Further, the result generated by the annotation can be converted into an encrypted character string and then stored, so that the security of the annotation data is ensured. After the whole labeling process is completed, the correctness of the data is confirmed again through the display process. Therefore, the embodiment of the invention provides an efficient gesture skeleton labeling method, which can provide better training data for a deep learning technology.
Fig. 8 is a schematic structural diagram of a key point labeling device according to an embodiment of the present invention, referring to fig. 8, the device includes:
The display module 801 is configured to display a labeling reference picture and a sample picture, where the labeling reference picture includes a plurality of first key points and number information of each first key point, and the number information of each first key point is used to indicate a labeling sequence for labeling key points;
The labeling module 802 is configured to label a plurality of second key points on the sample picture according to a plurality of triggering operations on the sample picture;
and the storage module 803 is configured to obtain a labeling result file of the sample picture based on the labeled second key points in the sample picture.
In some embodiments, the labeling module 802 is configured to:
When a triggering operation is detected on the sample picture, determining the number information of a second key point corresponding to the triggering operation according to the marked second key point;
and determining the coordinates of the contact point of the triggering operation as the coordinates of the second key point on the sample picture, and displaying the second key point and the number information of the second key point on the coordinates of the triggering point.
In some embodiments, the labeling module 802 is configured to:
When the triggering operation is detected on the sample picture, adding 1 to the maximum number information in the marked second key points to obtain the number information of the second key points corresponding to the triggering operation.
In some embodiments, the display module 801 is further configured to display at least one keypoint connection on the sample picture when displaying the plurality of second keypoints on the sample picture.
In some embodiments, the display module 801 is further configured to connect the second keypoints indicated by the preset connection relationship with a line according to the preset connection relationship between the plurality of second keypoints, so as to obtain the at least one keypoint line.
In some embodiments, the apparatus further comprises a first instruction processing module to:
When a first instruction is received, deleting the previous second key point and the number information of the previous second key point, wherein the first instruction is used for indicating to return to the previous key point for marking again; or alternatively, the first and second heat exchangers may be,
And when a second instruction is received, deleting all second key points in the sample picture and the number information of all second key points, wherein the second instruction is used for deleting all marked second key points.
In some embodiments, the apparatus further comprises a second instruction processing module for:
When a third instruction is received, switching and displaying the next sample picture, wherein the third instruction is used for switching the picture; or alternatively, the first and second heat exchangers may be,
When a fourth instruction is received, the labeling process is exited and unlabeled sample pictures are recorded, wherein the fourth instruction is used for exiting the labeling process; or alternatively, the first and second heat exchangers may be,
And when a fifth instruction is received, saving the marked sample picture and switching to the next sample picture, wherein the fifth instruction is used for the sample picture to finish marking.
In some embodiments, the storage module 803 is configured to encrypt coordinates of the plurality of second keypoints on the sample picture to obtain an encrypted string, and store the encrypted string as a labeling result file of the sample picture.
In some embodiments, the apparatus further comprises: a decryption module for:
when receiving a picture checking instruction, decrypting an encrypted character string in a labeling result file of the sample picture;
And displaying a plurality of second key points of the sample picture on the sample picture based on the decrypted data.
In some embodiments, the display module 801 is further configured to, when displaying the sample picture and the plurality of second keypoints on the sample picture, display a line connected by the plurality of second keypoints on the sample picture.
In some embodiments, the apparatus further comprises a third instruction processing module for:
When a sixth instruction is received in the process of displaying the sample picture, switching and displaying the next marked sample picture, wherein the sixth instruction is used for switching the marked picture; or alternatively, the first and second heat exchangers may be,
And deleting the labeling result file of the sample picture when a seventh instruction is received in the display process of the sample picture, wherein the seventh instruction is used for deleting the labeling result file of the sample picture.
Any combination of the above-mentioned optional solutions may be adopted to form an optional embodiment of the present disclosure, which is not described herein in detail.
It should be noted that: in the key point labeling device provided in the above embodiment, only the division of the above functional modules is used for illustration, and in practical application, the above functional allocation may be performed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules, so as to complete all or part of the functions described above. In addition, the key point labeling device provided in the above embodiment and the key point labeling method embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which is not repeated here.
Taking a specific implementation manner of the key point labeling device as an example, the device can include the following functional modules:
A constant module for representing the labeling of the sample pictures, the connection relation among the key points and the color of the connecting line; the constant module may be provided as part of the functionality in the display module to determine the specific display form of each point on display.
The module monitors the action of the mouse, when the left button of the mouse is pressed, the pressed position is recorded as a marking position;
The module is used for monitoring the action of the keyboard, and when a marker needs to return one-time marking, start marking from the head, skip pictures, temporarily exit the system or save marking results, the corresponding position on the keyboard is pressed down;
The module for detecting the motion can be provided as part of the function of a labeling module for monitoring the motion of the mouse so as to label, and monitoring the motion of the keyboard so as to modify the label.
The storage module can provide an encryption storage function for the storage module, converts the labeling result into a character string format, sets a key to be encrypted by a DES encryption algorithm, and then stores the key in a result file;
The decryption module decrypts the result by using the same key as the encryption module when reading the data from the result file;
The labeling display module records the current labeling result in the whole labeling time period and displays the current labeling result on an image interface interacted with a labeling person;
and the marking result display module reads data from the result file for decryption after marking is finished and displays the data on the image interface.
The marking display module and the marking result display module can be provided as part of functions of the display module and used for displaying information to be displayed.
The modules of the marking tool keep certain independence, and the marking tool is divided into different tasks, but the marking tool has certain mutual dependency relationship, and the dependency relationship is shown in the topological diagram of fig. 9. It can be seen from the figure that the labeling process needs the coordination of the constant module, the monitoring mouse action module and the monitoring keyboard action module, the display process needs the coordination of the constant module and the monitoring keyboard action module, and the coordinates of all key points are already determined, so that the process does not need the module for monitoring the mouse action.
Fig. 10 is a schematic structural diagram of a computer device according to an embodiment of the present invention, where the computer device 1000 may have a relatively large difference due to different configurations or performances, and may include one or more processors (central processing units, CPU) 1001 and one or more memories 1002, where the memories 1002 store at least one instruction, and the at least one instruction is loaded and executed by the processors 1001 to implement the methods provided in the foregoing method embodiments. Of course, the computer device may also have a wired or wireless network interface, a keyboard, an input/output interface, and other components for implementing the functions of the device, which are not described herein.
In an exemplary embodiment, a computer readable storage medium, such as a memory comprising instructions executable by a processor in a terminal to perform the method of keypoint labeling in the embodiments described below, is also provided. For example, the computer readable storage medium may be ROM, random Access Memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program for instructing relevant hardware, where the program may be stored in a computer readable storage medium, and the storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (9)
1. A method of key point labeling, performed by a computer device, the method comprising:
displaying an initial interface of an application program, wherein the initial interface comprises a plurality of function options, each function option corresponds to a picture marking function, and the picture marking function comprises a gesture key point marking function;
In response to receiving triggering operation of a gesture key point labeling function, displaying a labeling reference picture and a sample picture corresponding to the function option, wherein the labeling reference picture comprises a plurality of first key points and numbering information of each first key point, and the numbering information of each first key point is used for indicating a labeling sequence for conducting key point labeling;
Displaying operation prompt information, wherein the operation prompt information is used for indicating how to label the sample picture to be labeled according to the first key point displayed on the labeling reference picture;
Marking a plurality of second key points on the sample picture according to a plurality of triggering operations on the sample picture, connecting the second key points by lines based on a preset connection relation, wherein the preset connection relation is used for reflecting key points belonging to the same skeleton branch, key point connecting lines on the same skeleton branch are displayed by adopting the same color, key point connecting lines on different skeleton branches are displayed by adopting different colors, in the process of marking the second key points, each marked second key point is connected with a marked second key point with a preset skeleton connection relation by lines, and the marked second key points are displayed on the sample picture, the computer equipment comprises a touch screen, the triggering operation comprises touch operations on the touch screen, and the second key points are displayed on an original image layer on the sample picture in a floating layer mode; according to the skeleton joint positions of a plurality of first key points in the marked reference picture, triggering operation is carried out on the corresponding skeleton joint positions on the sample picture, the computer equipment determines the next key point of the marked second key point as the second key point corresponding to the triggering operation according to the detected triggering operation and the marked second key point, the triggering point coordinates of the triggering operation are determined as the coordinates of the second key point on the picture, and the computer equipment displays the second key point on an interface;
When a first instruction is received, deleting the previous second key point and the number information of the previous second key point, wherein the first instruction is used for indicating to return to the previous key point for marking again; when a second instruction is received, deleting all second key points in the sample picture and the number information of all second key points, wherein the second instruction is used for deleting all marked second key points;
When a third instruction is received, switching and displaying the next sample picture, wherein the third instruction is used for switching the picture; when a fourth instruction is received, the labeling process is exited and unlabeled sample pictures are recorded, wherein the fourth instruction is used for exiting the labeling process and directly displaying the unlabeled sample pictures when the application program is started next time; when a fifth instruction is received, the marked sample picture is saved and is switched to the next sample picture, wherein the fifth instruction is used for indicating that the sample picture is marked;
When a labeling completion instruction is received, encrypting based on coordinates of a plurality of labeled second key points in the sample picture to obtain an encrypted character string, and storing the encrypted character string as a labeling result file of the sample picture;
When a picture viewing instruction is received, decrypting the encrypted character string in the labeling result file of the sample picture, and displaying a plurality of second key points of the sample picture on the sample picture based on data obtained after decryption;
and deleting the labeling result file of the sample picture when a seventh instruction is received in the display process of the sample picture, wherein the seventh instruction is used for deleting the labeling result file of the sample picture.
2. The method of claim 1, wherein labeling a plurality of second keypoints on the sample picture according to a plurality of trigger operations on the sample picture comprises:
When a triggering operation is detected on the sample picture, determining the number information of a second key point corresponding to the triggering operation according to the marked second key point;
and determining the coordinates of the contact point of the triggering operation as the coordinates of the second key point on the sample picture, and displaying the second key point and the number information of the second key point on the coordinates of the triggering point.
3. The method according to claim 2, wherein when the triggering operation is detected on the sample picture, determining the number information of the second key point corresponding to the triggering operation according to the noted second key point includes:
When the triggering operation is detected on the sample picture, adding 1 to the maximum number information in the marked second key points to obtain the number information of the second key points corresponding to the triggering operation.
4. The method according to claim 1, wherein the method further comprises:
and displaying lines obtained by connecting the plurality of second key points on the sample picture when the sample picture and the plurality of second key points on the sample picture are displayed.
5. The method according to claim 1, wherein the method further comprises:
and when a sixth instruction is received in the process of displaying the sample picture, switching and displaying the next marked sample picture, wherein the sixth instruction is used for switching the marked picture.
6. A key point labeling apparatus, the apparatus comprising:
The display module is used for displaying an initial interface of the application program, the initial interface comprises a plurality of function options, each function option corresponds to a picture marking function, and the picture marking function comprises a gesture key point marking function; in response to receiving triggering operation of a gesture key point labeling function, displaying a labeling reference picture and a sample picture corresponding to the function option, wherein the labeling reference picture comprises a plurality of first key points and numbering information of each first key point, and the numbering information of each first key point is used for indicating a labeling sequence for conducting key point labeling; displaying operation prompt information, wherein the operation prompt information is used for indicating how to label the sample picture to be labeled according to the first key point displayed on the labeling reference picture;
The marking module is used for marking a plurality of second key points on the sample picture according to a plurality of triggering operations on the sample picture and connecting the second key points by lines based on a preset connection relation, the preset connection relation is used for reflecting key points belonging to the same skeleton branch, key point connecting lines on the same skeleton branch are displayed by adopting the same color, key point connecting lines on different skeleton branches are displayed by adopting different colors, each second key point is marked in the process of marking the second key points, the marked second key points are connected with marked second key points with lines with a preset skeleton connection relation, the marked second key points are displayed on the sample picture, the computer equipment comprises a touch screen, the triggering operation comprises touch operations on the touch screen, and the second key points are displayed on an original image layer on the sample picture in a floating layer mode; according to the skeleton joint positions of a plurality of first key points in the marked reference picture, triggering operation is carried out on the corresponding skeleton joint positions on the sample picture, the computer equipment determines the next key point of the marked second key point as the second key point corresponding to the triggering operation according to the detected triggering operation and the marked second key point, the triggering point coordinates of the triggering operation are determined as the coordinates of the second key point on the picture, and the computer equipment displays the second key point on an interface;
The first instruction processing module is used for deleting the previous second key point and the number information of the previous second key point when receiving a first instruction, wherein the first instruction is used for indicating to return to the previous key point for marking again; when a second instruction is received, deleting all second key points in the sample picture and the number information of all second key points, wherein the second instruction is used for deleting all marked second key points;
The second instruction processing module is used for switching and displaying the next sample picture when a third instruction is received, and the third instruction is used for switching the picture; when a fourth instruction is received, the labeling process is exited and unlabeled sample pictures are recorded, wherein the fourth instruction is used for exiting the labeling process and directly displaying the unlabeled sample pictures when the application program is started next time; when a fifth instruction is received, the marked sample picture is saved and is switched to the next sample picture, wherein the fifth instruction is used for indicating that the sample picture is marked;
the storage module is used for encrypting based on coordinates of a plurality of marked second key points in the sample picture on the sample picture when a marking completion instruction is received, so as to obtain an encrypted character string, and storing the encrypted character string as a marking result file of the sample picture;
the decryption module is used for decrypting the encrypted character strings in the labeling result file of the sample picture when receiving the picture viewing instruction, and displaying a plurality of second key points of the sample picture on the sample picture based on the data obtained after decryption;
and the third instruction processing module is used for deleting the labeling result file of the sample picture when a seventh instruction is received in the display process of the sample picture, wherein the seventh instruction is used for deleting the labeling result file of the sample picture.
7. The apparatus of claim 6, wherein the labeling module is configured to:
When a triggering operation is detected on the sample picture, determining the number information of a second key point corresponding to the triggering operation according to the marked second key point;
and determining the coordinates of the contact point of the triggering operation as the coordinates of the second key point on the sample picture, and displaying the second key point and the number information of the second key point on the coordinates of the triggering point.
8. A computer device comprising a processor and a memory having stored therein at least one instruction that is loaded and executed by the processor to implement the operations performed by the keypoint labeling method of any of claims 1-5.
9. A computer-readable storage medium having stored therein at least one instruction that is loaded and executed by a processor to implement the operations performed by the keypoint labeling method of any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810678758.9A CN110147701B (en) | 2018-06-27 | 2018-06-27 | Key point labeling method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810678758.9A CN110147701B (en) | 2018-06-27 | 2018-06-27 | Key point labeling method, device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110147701A CN110147701A (en) | 2019-08-20 |
| CN110147701B true CN110147701B (en) | 2024-08-16 |
Family
ID=67589123
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810678758.9A Active CN110147701B (en) | 2018-06-27 | 2018-06-27 | Key point labeling method, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110147701B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111259822A (en) * | 2020-01-19 | 2020-06-09 | 杭州微洱网络科技有限公司 | Method for detecting key point of special neck in E-commerce image |
| CN113539446A (en) * | 2020-04-21 | 2021-10-22 | 杭州普健医疗科技有限公司 | CT image labeling method and system, storage medium and terminal |
| CN111626233B (en) * | 2020-05-29 | 2021-07-13 | 江苏云从曦和人工智能有限公司 | Key point marking method, system, machine readable medium and equipment |
| US11361490B2 (en) | 2020-08-18 | 2022-06-14 | Here Global B.V. | Attention guidance for ground control labeling in street view imagery |
| CN112508127A (en) * | 2020-12-22 | 2021-03-16 | 北京百度网讯科技有限公司 | Data labeling method and device, electronic equipment, medium and product |
| CN112528929A (en) * | 2020-12-22 | 2021-03-19 | 北京百度网讯科技有限公司 | Data labeling method and device, electronic equipment, medium and product |
| CN114757250A (en) * | 2020-12-29 | 2022-07-15 | 华为云计算技术有限公司 | Image processing method and related equipment |
| CN113010069A (en) * | 2021-03-12 | 2021-06-22 | 浙江大华技术股份有限公司 | Switching method and device for picture labels, electronic device and storage medium |
| CN113011403B (en) * | 2021-04-30 | 2023-11-24 | 恒睿(重庆)人工智能技术研究院有限公司 | Gesture recognition method, system, medium and device |
| CN113407869B (en) * | 2021-06-02 | 2024-07-16 | 北京爱笔科技有限公司 | Beacon labeling method, device, computer equipment and storage medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103390282A (en) * | 2013-07-30 | 2013-11-13 | 百度在线网络技术（北京）有限公司 | Image tagging method and device |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5502803A (en) * | 1993-01-18 | 1996-03-26 | Sharp Kabushiki Kaisha | Information processing apparatus having a gesture editing function |
| US8396287B2 (en) * | 2009-05-15 | 2013-03-12 | Google Inc. | Landmarks from digital photo collections |
| EP3107032A1 (en) * | 2015-06-17 | 2016-12-21 | Thomson Licensing | Method and device for facial landmark annotation and corresponding computer program product and computer-readable medium |
| CN105184283A (en) * | 2015-10-16 | 2015-12-23 | 天津中科智能识别产业技术研究院有限公司 | Method and system for marking key points in human face images |
| US10852936B2 (en) * | 2016-09-23 | 2020-12-01 | Apple Inc. | Devices, methods, and graphical user interfaces for a unified annotation layer for annotating content displayed on a device |
-
2018
- 2018-06-27 CN CN201810678758.9A patent/CN110147701B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103390282A (en) * | 2013-07-30 | 2013-11-13 | 百度在线网络技术（北京）有限公司 | Image tagging method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110147701A (en) | 2019-08-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110147701B (en) | Key point labeling method, device, computer equipment and storage medium | |
| CN103176690B (en) | Display control apparatus, display control method and program | |
| CN109561271B (en) | Method for guiding terminal operation, first terminal and second terminal | |
| CN111832468A (en) | Gesture recognition method and device based on biological recognition, computer equipment and medium | |
| CN107666987A (en) | Robotic process automates | |
| CN104899825A (en) | Method and device for modeling picture figure | |
| WO2022174605A1 (en) | Gesture recognition method, gesture recognition apparatus, and smart device | |
| CN108229373A (en) | Calligraphy exercising method and calligraphy practicing device, computer equipment and storage medium | |
| CN109215441A (en) | A kind of Piano Teaching method, apparatus and computer storage medium | |
| CN113703631A (en) | Writing control method and device, electronic equipment and storage medium | |
| CN113672143B (en) | Image labeling method, system, device and storage medium | |
| US20110037731A1 (en) | Electronic device and operating method thereof | |
| TWI505173B (en) | System and method for improving recognition of a touch keyboard in an electronic device | |
| US20140354676A1 (en) | Hair colouring device and method | |
| CN111857499B (en) | Information prompting method and device | |
| CN110333780A (en) | Function triggering method, device, equipment and storage medium | |
| CN114449323B (en) | Video generation method and electronic device | |
| CN115134317B (en) | Message display method, device, storage medium and electronic device | |
| CN109471525A (en) | Business processing method, device, computer equipment and storage medium | |
| CN114245193B (en) | Display control method, device and electronic device | |
| JP2010198076A (en) | Handwriting authentication device, display method for handwriting authentication pattern and display program for handwriting authentication pattern | |
| CN113282768A (en) | Multimedia file processing method and device and electronic equipment | |
| JP7255619B2 (en) | Information processing device, information processing method, information processing program, and information processing system | |
| EP4510092A1 (en) | Palm contour extraction method and apparatus, and control instruction generation method and apparatus | |
| CN115396245B (en) | Content sharing method and device |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TG01 | Patent term adjustment | ||
| TG01 | Patent term adjustment |