CN109242868A - Image processing method, device, electronic equipment and storage medium - Google Patents
Image processing method, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN109242868A CN109242868A CN201811085280.5A CN201811085280A CN109242868A CN 109242868 A CN109242868 A CN 109242868A CN 201811085280 A CN201811085280 A CN 201811085280A CN 109242868 A CN109242868 A CN 109242868A
- Authority
- CN
- China
- Prior art keywords
- size
- image
- physical feeling
- profile
- skeleton
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30008—Bone
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the present application provides a kind of image processing method, device, electronic equipment and storage medium, is related to technical field of image processing.Method includes: the image for obtaining the physical feeling comprising target object.According to image and body skeleton detection model, the skeleton of physical feeling is determined;And according to image and body contour identification model, determine the profile of physical feeling.According to skeleton and profile, determine that the size of physical feeling is first size.The size of physical feeling is adjusted to the second size by first size.It solves the troublesome operation that user voluntarily adjusts in the prior art to standard stature it is thus achieved that automatically adjusting the stature of target object, and operates bad the technical issues of causing stature to be distorted.
Description
Technical field
This application involves technical field of image processing, set in particular to a kind of image processing method, device, electronics
Standby and storage medium.
Background technique
Currently, people like the photo upload of oneself to mention to social software with the extensive use of various social softwares
Share on the platform of confession with other people.
To show other people for the image of oneself is more beautiful by photo, people just need to come by third party software
The posture of oneself on photo is adjusted.But not only cumbersome, and user and bad are adjusted by third party software
Suitable adjustment degree is controlled, posture adjusted is caused to be distorted.
Summary of the invention
The application is to provide a kind of image processing method, device, electronic equipment and storage medium, on effectively improving
State technical problem.
To achieve the goals above, embodiments herein is accomplished in that
In a first aspect, the embodiment of the present application provides a kind of image processing method, which comprises
Obtain the image of the physical feeling comprising target object;
According to described image and body skeleton detection model, the skeleton of the physical feeling is determined;And according to described
Image and body contour identification model, determine the profile of the physical feeling;
According to the skeleton and the profile, determine that the size of the physical feeling is first size;
The size of the physical feeling is adjusted to the second size by the first size.
With reference to first aspect, the embodiment of the present application provides the first possible embodiment of first aspect, and described
According to the skeleton and the profile, determine that the size of the physical feeling is first size, comprising:
According to the skeleton and the profile, the fitted figure for being fitted the physical feeling is determined;
According to the first size of the fitted figure, the first size that the size of the physical feeling is is determined.
With reference to first aspect, the embodiment of the present application provides second of possible embodiment of first aspect, and described
According to the skeleton and the profile, the fitted figure for being fitted the physical feeling is determined, comprising:
According at least two articulation nodes of the skeleton, and according to multiple profile points of the profile, determine to intend
Close the fitted figure of at least two articulation nodes and the multiple profile point.
With reference to first aspect, the embodiment of the present application provides the third possible embodiment of first aspect,
The fitted figure is ellipse, and the first size indicates the long axis length or minor axis length of the ellipse;
Or
The fitted figure is circle, and the first size indicates the circular radius length;Or
The fitted figure is rectangle, and the first size indicates the side length of the rectangle.
With reference to first aspect, the embodiment of the present application provides the 4th kind of possible embodiment of first aspect, and described
According to the first image and body skeleton detection model, the skeleton of the physical feeling is determined, comprising:
According to described image and the body skeleton detection model, determine to be located at the physical feeling from described image
On at least two articulation nodes, at least two articulation nodes determine the skeleton of the physical feeling.
With reference to first aspect, the embodiment of the present application provides the 5th kind of possible embodiment of first aspect, and described
According to second image and body contour identification model, the profile of the physical feeling is determined, comprising:
According to described image and body contour identification model, determine to be located at the physical feeling edge from described image
Multiple profile points, the multiple profile point determines the profile of the physical feeling.
With reference to first aspect, any possible embodiment in the first of first aspect to the 5th kind, the application
Embodiment provides the 6th kind of possible embodiment of first aspect, described according to the skeleton and the profile, determines
Out the size of the physical feeling be first size after, the method also includes:
Judge whether the first size of the physical feeling matches with second size;
If it is not, executing step: the size of the physical feeling is adjusted to the second size by first size.
With reference to first aspect, the embodiment of the present application provides the 7th kind of possible embodiment of first aspect, described to incite somebody to action
The size of the physical feeling is adjusted to the second size by the first size, comprising:
According to second size, by the profile in described image by the corresponding position of the first size adjust to
The corresponding position of second size obtains the described image that the physical feeling is second size.
With reference to first aspect, any possible embodiment in the first of first aspect to the 5th kind, the application
Embodiment provides the 8th kind of possible embodiment of first aspect, the figure for obtaining the physical feeling comprising target object
Picture, comprising:
Obtain the original image of the physical feeling comprising target object;
Determine physical feeling ratio shared by the original image;
The original image is handled, obtaining the physical feeling, proportion is different from the ratio pre- in the picture
If the image of ratio.
Second aspect, the embodiment of the present application provide a kind of image processing apparatus, and described device includes:
Image obtains module, for obtaining the image of the physical feeling comprising target object;
First determining module, for determining the physical feeling according to described image and body skeleton detection model
Skeleton;And according to described image and body contour identification model, determine the profile of the physical feeling;
Second determining module, for according to the skeleton and the profile, determining that the size of the physical feeling is the
One size;
Module is adjusted, for the size of the physical feeling to be adjusted to the second size by the first size.
In conjunction with second aspect, the embodiment of the present application provides the first possible embodiment of second aspect,
Described second determining module is also used to determine to be fitted the body according to the skeleton and the profile
The fitted figure of position;According to the first size of the fitted figure, the first size that the size of the physical feeling is is determined.
In conjunction with second aspect, the embodiment of the present application provides second of possible embodiment of second aspect,
Second determining module is also used at least two articulation nodes according to the skeleton, and according to the wheel
Wide multiple profile points, determine the fitted figure for being fitted at least two articulation nodes and the multiple profile point.
In conjunction with second aspect, the embodiment of the present application provides the third possible embodiment of second aspect,
First determining module is also used to according to described image and the body skeleton detection model, from described image
In determine at least two articulation nodes being located in the physical feeling, at least two articulation nodes determine the body
The skeleton of body region.
In conjunction with second aspect, the embodiment of the present application provides the 5th kind of possible embodiment of second aspect,
First determining module is also used to according to described image and body contour identification model, from described image really
Multiple profile points positioned at the physical feeling edge are made, the multiple profile point determines the profile of the physical feeling.
In conjunction with second aspect, second aspect the first to any possible embodiment in the 4th kind, the application
Embodiment provides the 5th kind of possible embodiment of second aspect, described device further include:
Judgment module, for judging whether the first size of the physical feeling matches with second size;If
It is no, it executes step: the size of the physical feeling is adjusted to the second size by first size.
In conjunction with second aspect, the embodiment of the present application provides the 6th kind of possible embodiment of second aspect,
The adjustment module is also used to according to second size, by the profile in described image by described first
The corresponding position of size adjusts position corresponding to second size, obtains the physical feeling as the institute of second size
State image.
In conjunction with second aspect, second aspect the first to any possible embodiment in the 4th kind, the application
Embodiment provides the 7th kind of possible embodiment of second aspect,
Described image obtains module, is also used to obtain the original image of the physical feeling comprising target object;Determine institute
State physical feeling ratio shared by the original image;The original image is handled, obtains the physical feeling in the picture
Proportion is the image of the preset ratio different from the ratio.
The third aspect, the embodiment of the present application provide a kind of electronic equipment, and the electronic equipment includes: processor, storage
Device, bus and communication interface;The processor, the communication interface and memory are connected by the bus.
The memory, for storing program.
The processor, for the program by calling storage in the memory, to execute first aspect and first
Image processing method described in any implementation of aspect.
Fourth aspect, the embodiment of the present application provide a kind of calculating of non-volatile program code that can be performed with computer
The readable storage medium of machine, said program code make any realization side of the computer execution first aspect and first aspect
Image processing method described in formula.
The beneficial effect of the embodiment of the present application is:
Due to the skeleton and profile according to physical feeling on image, the size of the physical feeling can be determined, therefore can be with
Automatically the size of physical feeling is adjusted, it is thus achieved that automatically adjust the stature of target object to standard stature,
It solves the troublesome operation that user in the prior art voluntarily adjusts, and operates bad the technical issues of causing stature to be distorted.
To enable the above objects, features, and advantages of the application to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
Technical solution in ord to more clearly illustrate embodiments of the present application, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only some embodiments of the application, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 shows the structural block diagram of a kind of electronic equipment of the application first embodiment offer;
Fig. 2 shows a kind of first pass figures for image processing method that the application second embodiment provides;
A kind of image processing method first that Fig. 3 shows the offer of the application second embodiment handles the schematic diagram of scene;
Fig. 4 shows a kind of second flow chart of image processing method of the application second embodiment offer;
Fig. 5 shows a kind of schematic diagram of image processing method second processing scene of the application second embodiment offer;
Fig. 6 shows a kind of schematic diagram of image processing method third processing scene of the application second embodiment offer;
Fig. 7 shows a kind of structural block diagram of image processing apparatus of the application 3rd embodiment offer.
Specific embodiment
Below in conjunction with attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Ground description, it is clear that described embodiments are only a part of embodiments of the present application, instead of all the embodiments.Usually exist
The component of the embodiment of the present application described and illustrated in attached drawing can be arranged and be designed with a variety of different configurations herein.Cause
This, is not intended to limit claimed the application's to the detailed description of the embodiments herein provided in the accompanying drawings below
Range, but it is merely representative of the selected embodiment of the application.Based on embodiments herein, those skilled in the art not into
Row goes out every other embodiment obtained under the premise of creative work, shall fall in the protection scope of this application.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing.Term " first ", " the
Two " etc. are only used for distinguishing description, are not understood to indicate or imply relative importance.
First embodiment
Referring to Fig. 1, the embodiment of the present application provides a kind of electronic equipment 10, electronic equipment 10 can be terminal, such as
PC (personal computer, PC), tablet computer, smart phone, personal digital assistant (personal
Digital assistant, PDA) etc.;Alternatively, electronic equipment 10 or server, such as network server, database
Server, Cloud Server or the server set that is made of multiple child servers at etc..Certainly, the above-mentioned equipment enumerated is for just
In understanding the present embodiment, the restriction to the present embodiment should not be used as.
The electronic equipment 10 may include: image collecting device 11, memory 12, communication interface 13, bus 14 and processing
Device 15.Wherein, processor 15, communication interface 13 and memory 12 and image collecting device 11 can be connected by bus 14.Fig. 1
Shown in electronic equipment 10 component and structure be it is illustrative, and not restrictive, as needed, electronic equipment 10 can also
With other assemblies and structure.
Image collecting device 11 can be camera, image collecting device 11 can be used for photographic subjects object image or
Video allows electronic equipment 10 to obtain the image or video of target object.
Memory 12 may include high-speed random access memory (Random Access Memory RAM), it is also possible to also
Including non-labile memory (non-volatile memory), for example, at least two magnetic disk storages.In the present embodiment,
Memory 11 stores program required for executing image processing method.
Bus 14 can be isa bus, pci bus or eisa bus etc..It is total that bus can be divided into address bus, data
Line, control bus etc..Only to be indicated with a four-headed arrow in Fig. 1, it is not intended that an only bus or one convenient for indicating
The other bus of type.
Processor 15 may be a kind of processing capacity IC chip with signal.During realization, above-mentioned side
Each step of method can be completed by the integrated logic circuit of the hardware in processor 15 or the instruction of software form.Above-mentioned
Processor 15 can be general processor, including central processing unit (Central Processing Unit, abbreviation CPU), network
Processor (Network Processor, abbreviation NP) etc.;It can also be digital signal processor (DSP), specific integrated circuit
(ASIC), ready-made programmable gate array (FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components.General processor can be microprocessor or the processor is also possible to any conventional processor etc..
The step of method in conjunction with disclosed in the embodiment of the present application, can be embodied directly in hardware decoding processor and execute completion, Huo Zheyong
Hardware and software module combination in decoding processor execute completion.Software module can be located at random access memory, flash memory, read-only
The storage medium of this fields such as memory, programmable read only memory or electrically erasable programmable memory, register maturation
In.
Method performed by the device of stream process or definition that the embodiment of the present application any embodiment discloses can be applied to
In processor 15, or realized by processor 15.Processor 15 is stored in after receiving and executing instruction by the calling of bus 15
After program in memory 11, processor 14 calls the 11 photographic subjects object of image collecting device stored by bus 13
Image or video can then execute the process of image processing method.
In addition, the detailed execution process about image processing method, will be described in detail following.
Second embodiment
Present embodiments provide a kind of image processing method, it should be noted that step shown in the flowchart of the accompanying drawings
It can execute in a computer system such as a set of computer executable instructions, although also, showing patrol in flow charts
Sequence is collected, but in some cases, it can be with the steps shown or described are performed in an order that is different from the one herein.Below to this
Embodiment describes in detail.
Referring to Fig. 2, the image processing method can be to be set by electronics in image processing method provided in this embodiment
Standby to execute, which includes: step S110, step S120, step S130 and step S140.
Step S110: the image of the physical feeling comprising target object is obtained.
Step S120: according to described image and body skeleton detection model, the skeleton of the physical feeling is determined;And
According to described image and body contour identification model, the profile of the physical feeling is determined.
Step S130: according to the skeleton and the profile, determine that the size of the physical feeling is first size.
Step S140: the size of the physical feeling is adjusted to the second size by the first size.
Below in conjunction with Fig. 2 to Fig. 6, the present processes process is described in detail.
Before step S110, electronic equipment can first carry out the training of model, to obtain trained body skeleton
Detection model and body contour identification model.
Before training, electronic equipment can be obtained for training body skeleton detection model and body contour identification model
Training sample image collection.Wherein, the mode that electronic equipment obtains training sample image collection can be with are as follows: electronic equipment be stored with
When the exterior storage medium connection of training sample image collection, the storage operation for responding user obtains training sample from storage medium
Image set;Or electronic equipment obtain the mode of training sample image collection can be with are as follows: respond the down operation of user from being stored with
Training sample image collection is obtained on the server or database of training sample image collection.
Training sample image collection can be to be made of, for example, the number of training sample image can multiple training sample images
Think tens of thousands of, and, it may include physical feeling in every training sample image in multiple training sample images.
Optionally, every training sample image comprising physical feeling can be to include some physical feeling on body
Image, can be understood as training sample image is to the image of some physical feeling feature, such as training sample image
It may include the positions such as head, shoulder, waist or crotch.Alternatively, every training sample image comprising physical feeling can also be with
Image comprising the body position on body, can be understood as training sample image is the image comprising half body portrait,
Such as training sample image may include the partial portion in head, shoulder, waist, crotch, forearm, large arm, shank and thigh.
Alternatively, every training sample image comprising physical feeling can also be the image comprising whole physical feelings on body,
Can be understood as training sample image be the image comprising whole body portrait, such as training sample image may include head, shoulder,
Whole positions in waist, crotch, forearm, large arm, shank and thigh.
As training body skeleton detection model and body contour identification model a kind of, optionally mode, electronic equipment can
To be trained respectively to body skeleton detection model and body contour identification model.
Electronic equipment can use training sample image collection and be trained to first nerves network, and utilize training sample
Image set is trained nervus opticus network.It wherein, can be with to the training process of first nerves network are as follows: first nerves network
The study of skeleton detection is carried out, by the skeleton to physical feeling in every training sample image to learn and determine to detect bone
The various parameters of the rule of frame and the rule.And may be to the training process of nervus opticus network: nervus opticus net
Network carries out the study of outline identification by the profile to physical feeling in every training sample image, is identified with learning and determining
The various parameters of the rule of profile and the rule.
After electronic equipment is trained first nerves network and nervus opticus network by training sample image collection, to protect
The subsequent accuracy in detection for obtaining body skeleton detection model is demonstrate,proved, it can also be using verifying sample graph image set to first after training
Nervus opticus network after neural network and training is verified.Wherein, verifying sample graph image set may be by multiple verifyings
Sample image composition for example, the number of verifying sample image can be nearly ten thousand, and verifies sample image and training sample image
The type for the content for including is consistent, i.e., also includes physical feeling on verifying sample image.
In the present embodiment, the process of verifying can pass through for the first nerves network after training to every verifying sample image
Skeleton detection is carried out, to adjust the rule and the rule of detection skeleton according to the testing result of sample image is verified to every
Various parameters, with improve identification accuracy rate.And the process of verifying can also pass through for the nervus opticus network after training
Outline identification is carried out to every verifying sample image, to adjust identification wheel according to the testing result of sample image is verified to every
The various parameters of wide rule and the rule, to improve the accuracy rate of identification.
After verifying first nerves network and nervus opticus network, electronic equipment can also pass through new training sample image
Collection is again trained the first nerves network after verifying and the nervus opticus network after verifying again, to be formed to training and be tested
The alternating poll of card executes.It is understood that each time poll training when training sample image collection image can with it is other
The image of training sample image collection when poll training is not identical, and verifying sample graph image set when poll training each time
Image can also be not identical as the image of the verifying sample graph image set when training of other polls, to guarantee training and testing each time
The validity of card.
In embodiment itself, during execution of the electronic equipment to the alternating poll of training and verifying, first nerves
Network and nervus opticus network can be higher and higher to the skeleton detection precision of training sample image in training;And first
Neural network and nervus opticus network, also can be higher and higher to the precision of the skeleton detection of verifying sample image in verifying.
So first nerves network is in training or verifying, when reaching preset threshold to skeleton detection precision, and nervus opticus network
In training or verifying, when also reaching preset threshold to the precision of outline identification, it is believed that first nerves network and second
Neural network is up to standard, thus electronic equipment can deconditioning and verifying, obtaining should be by first nerves network training and testing
Obtained body skeleton detection model is demonstrate,proved, and obtains this and is identified by the body contour that nervus opticus network training and verifying obtain
Model.
It optionally, is the accuracy convenient for clearly knowing body skeleton detection model and body contour identification model, electricity
Sub- equipment can also carry out accuracy to body skeleton detection model and body contour identification model using test sample image collection
Test.Wherein, test sample image collection may be and be made of multiple test sample images, for example, of test sample image
Number or nearly ten thousand, and the type of test sample image and the training sample image content that includes is consistent, i.e. test specimens
It also include physical feeling on this image.In the present embodiment, the process of test can be body skeleton detection model to test sample
Every test sample image of image set carries out skeleton detection, and obtains and carry out the accurate of skeleton detection to test sample image collection
Degree.Certainly, the process of test may be every test sample image of the body contour identification model to test sample image collection
Outline identification is carried out, and obtains the accuracy for carrying out outline identification to test sample image collection.Under normal conditions, by training and
Body skeleton detection model and body contour identification model after verifying, body skeleton detection model and body contour identification model
Accuracy obtained by being tested by test sample image collection is 99% or more.
As another optionally mode of training body skeleton detection model and body contour identification model, electronic equipment
Body skeleton detection model and body contour identification model can be trained together.
First nerves network and nervus opticus network can be fitted in a neural network, and electronic equipment can then pass through
Training sample image collection is trained and verifies to the neural network with verifying sample graph image set training in rotation, obtains body bone in this way
The detection model of frame detection model and the fitting of body contour identification model.So, then by test sample image collection to the detection
Model carries out the test of accuracy, and the training of the detection model is just completed.
It is understood that the detailed training process of the detection model is seen documented by a kind of optionally mode above-mentioned
Content is just not repeated herein.
It will also be appreciated that body skeleton detection model and body contour identification model are trained respectively in use
When first way, image is inputted into body skeleton detection model, then body skeleton detection model can be exported to detection
Skeleton result;And image is inputted into body contour identification model again, then body contour identification model can be exported to wheel
Wide recognition result.But using second be trained together to body skeleton detection model and body contour identification model
When mode, image is inputted into the detection model that the training obtains and is handled, which can export respectively and detect
Skeleton and the profile that identifies.
It will also be appreciated that in actual use, it can be according to user's using first way or the second way
Business demand, to this present embodiment and without limitation.But the present embodiment can in the first manner for carry out scheme it is detailed
Describe in detail it is bright, in order to understand this programme.
After electronic equipment is completed to training and obtaining body skeleton detection model and body contour identification model, in subsequent use
Family is related to the processing to the image of target object in the use process of electronic equipment, then electronic equipment can then execute
Step S110.
Step S110: the image and image of the physical feeling comprising target object are obtained.
If user's determination needs to optimize the stature of target object in an image or a video, user can set in electronics
The opening operation of standby upper execution stature optimization function, wherein target object can also be with for user oneself or target object
For except other objects with open air, other objects can be for such as other people, animal.Electronic equipment can be according to detecting
The opening operation, so that it is determined that can be optimized to the stature of target object in image or video flowing.Wherein, image or video
Stream can be to acquire to obtain in the on state by the image collecting device on electronic equipment, and stores to it;Or
Person, image or video flowing may be to obtain and store from other equipment, for example, obtaining by network from cloud, pass through data
Line is obtained from the other equipment connecting with electronic equipment.
If needing to optimize in electronic equipment is image, electronic equipment can be obtained from storage region comprising target object
Physical feeling original image.
If needing to optimize in electronic equipment is video flowing, wherein every frame image in video can include target pair
The physical feeling of elephant.So, electronic equipment can extract the body comprising target object with the preview video flowing from the video flowing
The original image of body region.
In the present embodiment, the mode that electronic equipment extracts original image from video flowing can be with are as follows: to guarantee at image
The quality of reason, in the case where the performance of electronic equipment meets, electronic equipment can be by the way of handling to video flowing one by one
In every frame image handled.So electronic equipment can be after the completion of handling previous frame image, according in video flowing
The sequence of every frame image extracts the original image of previous frame image consecutive frame.
In the present embodiment, the mode that electronic equipment extracts original image from video flowing may be: because of video flowing
The stature of middle target object is not in transition, therefore the content in two images of several frames or more than ten frames being separated by video flowing
Difference is simultaneously little, then electronic equipment can using take out frame processing by the way of to the every frame image being drawn into video flowing at
Reason.So electronic equipment reaches preset time point after the completion of to the preceding primary image procossing for taking out frame, then takes out from video flowing
Take out the original image of current preview extremely.
It is understood that setting preset time point make electronic equipment to it is preceding it is primary take out frame image procossing after the completion of again
After have passed through one section of duration, electronic equipment can extract original image again.So in this section of duration, electronic equipment is
Through several frames of continuous preview in video streaming or more than ten frame images, thus the original image extracted later just with preceding primary pumping
Several frames or more than ten frames have been differed between the image of frame.
In the present embodiment, since the stature of user in video flowing is not in transition, therefore electronic equipment can also be taken out
Two original images are taken out to be handled, that is, extract the first original image and the second original image.To the first original image
It subsequent will be illustrated with the processing mode of the second original image.It is understood that the first original image and the second original graph
As that can may be among video flowing for two field pictures or the first original image and the second original image adjacent in video flowing
Every the two field pictures that several frames or more than ten frames acquire respectively.
Since in video flowing or image taking, distance of the target object apart from seat in the plane might not meet gauged distance,
It may cause ratio of the physical feeling shared by image and not up to preset ratio, wherein preset ratio can be used as electronics
The reference standard that equipment handles physical feeling in image, the ratio of physical feeling all reaches the ginseng in every image
Standard is examined, then electronic equipment differentiation can occur to avoid the processing result to physical feeling in multiple images.
Therefore after obtaining original graph, electronic equipment can determine that physical feeling exists by the analysis to original image
Ratio shared by original image simultaneously judges whether ratio is identical as preset ratio.
When being judged as that ratio is identical as preset ratio, electronic equipment does not just have to carry out original graph at cutting or encirclement
Reason, and using original image as image used in subsequent processing.
When being judged as that ratio and preset ratio be not identical, electronic equipment can be cut or be surrounded to original image
Processing, so that ratio is increased or reduces and reaches preset ratio.In this way, electronic equipment is obtained with physical feeling in image
Middle proportion is the image of the preset ratio different from ratio.
After obtaining image, electronic equipment can continue to execute step S200.
It should be noted that in the case where original image is opened in acquisition two, electronic equipment can to the first original image into
Row is same as the above-mentioned processing to original image, to obtain the first image by the first original image;Certainly, electronic equipment can be with
Second original image is carried out to be same as the above-mentioned processing to original image, to obtain the second image by the second original image.
Step S120: according to described image and body skeleton detection model, the skeleton of the physical feeling is determined;And
According to described image and body contour identification model, the profile of the physical feeling is determined.
Electronic equipment can call the preset body skeleton detection model, and the first image can be input to body bone
In frame detection model.Default rule is based on according to body skeleton detection model and each parameter analyzes image, and electronics is set
It is standby to obtain the result exported after the analysis of body skeleton detection model.The result can be to determine to be located at body from image
At least two articulation nodes on body region.Wherein it is determined that at least two articulation nodes can be with out are as follows: determine that this at least two is closed
Successively point in each articulation nodes coordinate, determine each articulation nodes be which joint and determine at least two joints section
Relationship in point between every two articulation nodes.
It is understood that electronic equipment can not have to carry out line by least two articulation nodes that will be determined come real
Construct the skeleton of the physical feeling in border.After determining at least two articulation nodes, due to each articulation nodes of acquisition
Coordinate, each articulation nodes be that relationship between which joint and every two articulation nodes is to constitute the physical feeling
Information needed for skeleton, therefore the coordinate of each articulation nodes, each articulation nodes are obtained as which joint and every two joint
Relationship between node is that is, determine the skeleton of the physical feeling.
Electronic equipment can also call the preset body contour identification model, and can input an image into body contour
In identification model.Default rule is based on according to body contour identification model and each parameter analyzes image, electronic equipment
The result exported after the analysis of body contour identification model can be obtained.The result can be to determine to be located at body from image
Multiple profile points of adjacent margins.Wherein it is determined that multiple profile points can be with out are as follows: determine each profile in multiple profile point
Point coordinate and determine each profile point be which physical feeling.
It will also be appreciated that electronic equipment can not have to by the way that the multiple profile points determined are carried out the practical structure of line
Building a figure indicates the profile of the physical feeling.After determining multiple profile point, due to each profile of acquisition
The coordinate and each profile point of point are which physical feeling is information needed for constituting the profile of the physical feeling, therefore is obtained each
The coordinate of profile point and each profile point are which physical feeling is that information needed for constituting the profile of the physical feeling is i.e. suitable
In the profile for determining the physical feeling.
As shown in Figure 3, it is assumed that the acquisition figure of electronic equipment is image X in Fig. 3, and image X contains the complete of target object
Body picture contains the physical feeling of target object whole in i.e. image X.By image X export respectively body skeleton detection model and
Body contour identification model can then obtain each body being made of in image X multiple articulation nodes A (being open circles in Fig. 3)
The skeleton at position, and by the profile of multiple profile point B (being filled circles in Fig. 3) each physical feeling constituted.Wherein, in Fig. 3
By the form that multiple articulation nodes A connections are more intuitively to embody skeleton, in order to it will be appreciated by those skilled in the art that but electricity
Connecting joint node this operation is not carried out in sub- equipment actual treatment.
It is also required to explanation, in the case where handling the first image and the second image, electronic equipment can be with
It is same as the above-mentioned processing to image to the first image is carried out, to determine the skeleton of physical feeling by the first image;Certainly, electric
Sub- equipment can also be same as the above-mentioned processing to image to the second image is carried out, to determine physical feeling by the second image
Profile.After determining skeleton and profile, electronic equipment can continue to execute step S300.
Step S130: according to the skeleton and the profile, determine that the size of the physical feeling is first size.
As shown in figure 4, step S130 may include: step S131 and step S132.
Step S131: according to the skeleton and the profile, the fitted figure for being fitted the physical feeling is determined.
Step S132: according to the first size of the fitted figure, the size of the physical feeling is first is determined
Size.
The method flow of the step S300 to the application is described in detail below.
Step S131: according to the skeleton and the profile, the fitted figure for being fitted the physical feeling is determined.For
During physical feeling determines the fitted figure for being fitted the physical feeling, electronic equipment can be according at least two passes of skeleton
Relationship in point between the coordinate and every two articulation nodes of each articulation nodes successively, and multiple profile points according to profile
In each profile point coordinate, electronic equipment can construct the fitted figure being fitted in the physical feeling.Wherein, it is fitted
The fitting of the figure physical feeling can be tangent for the part edge and profile of fitted figure.
In the present embodiment, according to the difference of physical feeling, fitted figure used by electronic equipment is also different.For example, body
Body region is waist, then the figure of the fitted figure waist can be oval or circle, in another example, physical feeling is small
Leg, then the figure of the fitted figure shank can be rectangle.
It is also required to explanation, in the case where handling the first image and the second image, electronic equipment can be with
It is mapped to the skeleton for handling the first image physical feeling obtained is corresponding in the second image, and can will handle second
The profile of image physical feeling obtained is corresponding to be mapped in the first image.In this way, electronic equipment can be based on the first figure
The profile of the skeleton and mapping determined on picture is the physical feeling fitted figure in the first image;And electronic equipment may be used also
It with the skeleton again based on the profile determined and mapping on the second image, then is the physical feeling fitted figure in the second image.
Continue above-mentioned it is assumed that referring to Fig. 5, Fig. 5 is to be illustrated for physical feeling is waist.In Fig. 5, according to
Articulation nodes A1, profile point B1 and profile point B2, electronic equipment can determine the fitted figure C for being fitted the waist, wherein quasi-
Closing figure C is ellipse.After obtaining fitted figure, electronic equipment can continue to execute step S320.
Step S132: according to the first size of the fitted figure, the size of the physical feeling is first is determined
Size.
Electronic equipment can according to the fitted figure fitted why the figure of type and dividing to fitted figure
Analysis, the size that electronic equipment can obtain the fitted figure is first size.And since the fitted figure is to be fitted the body
Body region determines the body then the first size of the fitted figure can be determined as the size of physical feeling by electronic equipment
The size of body region is first size.
In the present embodiment, since the type of fitted figure can be different, therefore the mode that first size indicates be can also be different.
Optionally, the type of fitted figure is ellipse, then first size indicates the long axis length or minor axis length of ellipse;Or fitting
The type of figure is circle, then first size indicates circular diameter length;Or the type of fitted figure is rectangle, then the first ruler
The very little side length for indicating rectangle.
It should be noted that if the type of fitted figure is ellipse, and the fitted figure is laterally disposed ellipse
(such as fitted figure is fitted waist), then such case, which typicallys represent, which can be adjusted elliptical long axis, (adjusts waist
Thickness), therefore first size indicate ellipse long axis length.And if the type of fitted figure is ellipse, and the fitted figure
Shape is placed longitudinally ellipse (such as fitted figure is fitted thigh), then typically represent can be to elliptical length for such case
Axis or short axle are adjusted (thickness and length of adjustment thigh), therefore first size indicates that the long axis length of ellipse or short axle are long
Degree.
Continue above-mentioned it is assumed that referring to Fig. 5, electronic equipment can determine fitting by analyzing fitted figure C
Figure C is laterally disposed ellipse, therefore electronic equipment can determine the first end point C1 and second on the long axis of fitted figure C
Endpoint C2 determines that the first size of fitted figure C is long axis length C12.
It is also required to explanation, in the case where electronic equipment handles the first image and the second image, then
While electronic equipment can determine fitted figure on first image and determine the first size of fitted figure, electricity
Sub- equipment can also determine fitted figure on second image and determine the first size of fitted figure.
Preset in the present embodiment, in electronic equipment physical feeling should adjust to the second size, which can
Think that preset default size or second size may be that user presets in the electronic device according to demand in electronic equipment
Size.For example, user it is expected waist more carefully, then the second size that physical feeling is waist can be arranged small by one
Point;Also for example, user it is expected leg more slightly, then the second size that physical feeling is leg can be arranged it is big a bit.
Correspondingly, electronic equipment obtain physical feeling first size after, electronic equipment it may determine that physical feeling the first ruler
It is very little whether to be matched with preset second size.
It is judged as that the first size of physical feeling is matched with preset second size in electronic equipment, i.e. the number of first size
When value numerical value identical as the numerical value of the second size or first size is differed within a preset range with the numerical value of the second size, then
Electronic equipment can be confirmed that the size of the physical feeling is user's desired size, and so there is no need to be adjusted to the first size.
Conversely, when electronic equipment is judged as that the first size of physical feeling and preset second size mismatch, then
Electronic equipment can be confirmed that the size of the physical feeling is not user's desired size, therefore the first size can be adjusted.
In the present embodiment, after electronic equipment confirms adjustable first size, electronic equipment can also continue to execute step
S140。
Step S140: the size of the physical feeling is adjusted to the second size by first size.
Since electronic equipment obtains the coordinate of each profile point, then electronic equipment can know the body on image
The profile of position is in the corresponding position of first size.Electronic equipment can be according to the second size, and redefining out fitting on image should
The coordinate of each new profile point in multiple new profile points of the new physical feeling of second size, wherein each new wheel
The coordinate of exterior feature point can represent the profile of the physical feeling on image in the corresponding position of the second size.Therefore electronic equipment root
According to the coordinate of each new profile point, electronic equipment can with the Coordinate Adjusting of each profile point in multiple profile points be with it is corresponding
The coordinate of each new profile point is identical, that is, realizes and adjust the profile on image to second by the corresponding position of first size
The corresponding position of size, and obtain the image that physical feeling is the second size.
It is also required to explanation, in the case where handling the first image and the second image, electronic equipment can be with
According to the second size, profile on the first image is adjusted into position corresponding to the second size by the corresponding position of first size, with
Obtain the first image that physical feeling is the second size.And/or
Electronic equipment can also be adjusted the profile on the second image by the corresponding position of first size according to the second size
To the corresponding position of the second size, to obtain the second image of the second size of physical feeling.
In the present embodiment, the physical feeling of the physical feeling of the first image and the second image is adjusted to the second size respectively
Detailed process can be just not repeated herein refering to aforementioned.
It please refers to Fig. 5 and Fig. 6, continues above-mentioned it is assumed that electronic equipment long axis length C12, which is determined, is less than pre-set dimension C34,
So electronic equipment redefines out the long axis that first end point C3 on long axis and the second endpoint C4 is constituted according to pre-set dimension C34
The identical fitted figure C ' of length pre-set dimension C34.Obtain fitted figure C ' can for fitted figure C first end point C1 to
Articulation nodes A1 direction movement and the second endpoint C2 also moved towards the direction of articulation nodes A1, so that it is quasi- to obtain this
Close figure C '.Therefore electronic equipment can determine profile point B1 and profile point B2 the case where being fitted fitted figure C ', profile
The new coordinate of point B1 and the new coordinate of profile point B2.So electronic equipment is according to the new coordinate and profile point of profile point B1
The Coordinate Adjusting of profile point B1 in Fig. 5 is new coordinate by the new coordinate of B2, and by the coordinate tune of profile point B2 in Fig. 5
It is whole be new coordinate, thus obtain the profile point B1 and profile point B2 in Fig. 6, then according in Fig. 6 profile point B1 and
The waist that profile point B2 obtains user in the new image X1 then becomes more very thin.
It will be appreciated that the second size can be greater than first size and the second size might be less that the first ruler
It is very little.User can come according to their own needs so that electronic equipment adjusts the stature of target object in image in this way.If second
Size is greater than first size, then the hero that the stature of partially thin target object in the picture can be adjusted is some;If second
Size is less than first size, then the stature of partially fat target object in the picture can be adjusted slenderrer.
It is understood that if electronic equipment be by the way of handling each frame image in video flowing, that
Electronic equipment shows the image that each frame physical feeling is the second size, to realize the broadcasting of video flowing, thus play
User can be clearly seen that the stature of target object in screen stream is oneself desired stature in screen stream.If but electronics is set
It is standby to be handled using the image to pumping frame in video flowing by the way of, then electronic equipment can according to will take out the image of frame by
The mode that first size is adjusted to the second size does corresponding adjustment to the image of other frames in screen stream and (no longer needs to detect it
Whether the image of its frame, which needs, adjusts), so that electronic equipment can also obtain the image that each frame physical feeling is the second size,
And also the image that each frame physical feeling is the second size is shown.
3rd embodiment
Referring to Fig. 7, the embodiment of the present application provides a kind of image processing apparatus 100, which can be with
Applied to electronic equipment, which includes:
Image obtains module 110, for obtaining the image of the physical feeling comprising target object.
First determining module 120, for determining the physical feeling according to described image and body skeleton detection model
Skeleton;And according to described image and body contour identification model, determine the profile of the physical feeling.
Second determining module 130, for determining that the size of the physical feeling is according to the skeleton and the profile
First size.
Module 140 is adjusted, for the size of the physical feeling to be adjusted to the second size by the first size.
It should be noted that due to it is apparent to those skilled in the art that, for the convenience and letter of description
Clean, system, the specific work process of device and unit of foregoing description can be with reference to corresponding in preceding method embodiment
Journey, details are not described herein.
It should be understood by those skilled in the art that, the embodiment of the present application can provide as the production of method, system or computer program
Product.Therefore, in terms of the embodiment of the present application can be used complete hardware embodiment, complete software embodiment or combine software and hardware
Embodiment form.Moreover, it wherein includes computer available programs generation that the embodiment of the present application, which can be used in one or more,
The meter implemented in the computer-usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) of code
The form of calculation machine program product.
In conclusion the embodiment of the present application provides a kind of image processing method, device, electronic equipment and storage medium.
Method includes: the image for obtaining the physical feeling comprising target object.According to image and body skeleton detection model, class origin is determined
The skeleton of body region;And according to image and body contour identification model, determine the profile of physical feeling.According to skeleton and wheel
Exterior feature determines that the size of physical feeling is first size.The size of physical feeling is adjusted to the second size by first size.
Due to the skeleton and profile according to physical feeling on image, the size of the physical feeling can be determined, therefore can be with
Automatically the size of physical feeling is adjusted, it is thus achieved that automatically adjust the stature of target object to standard stature,
It solves the troublesome operation that user in the prior art voluntarily adjusts, and operates bad the technical issues of causing stature to be distorted.
The above is only preferred embodiment of the present application, are not intended to limit this application, for those skilled in the art
For member, various changes and changes are possible in this application.Within the spirit and principles of this application, it is made it is any modification,
Equivalent replacement, improvement etc., should be included within the scope of protection of this application.It should also be noted that similar label and letter are under
Similar terms are indicated in the attached drawing in face, therefore, once being defined in a certain Xiang Yi attached drawing, are not then needed in subsequent attached drawing
It is further defined and explained.
More than, the only specific embodiment of the application, but the protection scope of the application is not limited thereto, and it is any to be familiar with
Those skilled in the art within the technical scope of the present application, can easily think of the change or the replacement, and should all cover
Within the protection scope of the application.Therefore, the protection scope of the application should be subject to the protection scope in claims.
Claims (12)
1. a kind of image processing method, which is characterized in that the described method includes:
Obtain the image of the physical feeling comprising target object;
According to described image and body skeleton detection model, the skeleton of the physical feeling is determined;And according to described image
With body contour identification model, the profile of the physical feeling is determined;
According to the skeleton and the profile, determine that the size of the physical feeling is first size;
The size of the physical feeling is adjusted to the second size by the first size.
2. image processing method according to claim 1, which is characterized in that it is described according to the skeleton and the profile,
The size for determining the physical feeling is first size, comprising:
According to the skeleton and the profile, the fitted figure for being fitted the physical feeling is determined;
According to the first size of the fitted figure, the first size that the size of the physical feeling is is determined.
3. image processing method according to claim 2, which is characterized in that it is described according to the skeleton and the profile,
Determine the fitted figure for being fitted the physical feeling, comprising:
According at least two articulation nodes of the skeleton, and according to multiple profile points of the profile, determine fitting institute
State the fitted figure of at least two articulation nodes and the multiple profile point.
4. image processing method according to claim 3, which is characterized in that
The fitted figure is ellipse, and the first size indicates the long axis length or minor axis length of the ellipse;Or
The fitted figure is circle, and the first size indicates the circular radius length;Or
The fitted figure is rectangle, and the first size indicates the side length of the rectangle.
5. image processing method according to claim 1, which is characterized in that described to be examined according to described image and body skeleton
Model is surveyed, determines the skeleton of the physical feeling, comprising:
According to described image and the body skeleton detection model, from determining to be located in the physical feeling in described image
At least two articulation nodes, at least two articulation nodes determine the skeleton of the physical feeling.
6. image processing method according to claim 1, which is characterized in that described to be known according to described image and body contour
Other model, determines the profile of the physical feeling, comprising:
According to described image and body contour identification model, determined from described image positioned at the more of the physical feeling edge
A profile point, the multiple profile point determine the profile of the physical feeling.
7. image processing method described in -6 any claims according to claim 1, which is characterized in that the method also includes:
Judge whether the first size of the physical feeling matches with second size;
If it is not, executing step: the size of the physical feeling is adjusted to the second size by first size.
8. image processing method according to claim 7, which is characterized in that the size by the physical feeling is by institute
It states first size and is adjusted to the second size, comprising:
According to second size, the profile in described image is adjusted by the corresponding position of the first size to described
The corresponding position of second size obtains the described image that the physical feeling is second size.
9. image processing method described in -6 any claims according to claim 1, which is characterized in that described obtain includes target pair
The image of the physical feeling of elephant, comprising:
Obtain the original image of the physical feeling comprising target object;
Determine physical feeling ratio shared by the original image;
The original image is handled, obtaining the physical feeling, proportion is the default ratio different from the ratio in the picture
The image of example.
10. a kind of image processing apparatus, which is characterized in that described device includes:
Image obtains module, for obtaining the image of the physical feeling comprising target object;
First determining module, for determining the skeleton of the physical feeling according to described image and body skeleton detection model;
And according to described image and body contour identification model, determine the profile of the physical feeling;
Second determining module, for determining that the size of the physical feeling is the first ruler according to the skeleton and the profile
It is very little;
Module is adjusted, for the size of the physical feeling to be adjusted to the second size by the first size.
11. a kind of electronic equipment, which is characterized in that the electronic equipment includes: processor, memory, bus and communication interface;
The processor, the communication interface and memory are connected by the bus;
The memory, for storing program;
The processor, for the program by calling storage in the memory, to execute such as any power of claim 1-9
Image processing method described in.
12. a kind of computer-readable storage media for the non-volatile program code that can be performed with computer, which is characterized in that institute
Stating program code makes the computer execute the image processing method as described in any claim of claim 1-9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811085280.5A CN109242868B (en) | 2018-09-17 | 2018-09-17 | Image processing method, image processing device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811085280.5A CN109242868B (en) | 2018-09-17 | 2018-09-17 | Image processing method, image processing device, electronic equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109242868A true CN109242868A (en) | 2019-01-18 |
CN109242868B CN109242868B (en) | 2021-05-04 |
Family
ID=65059610
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811085280.5A Active CN109242868B (en) | 2018-09-17 | 2018-09-17 | Image processing method, image processing device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109242868B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109683084A (en) * | 2019-01-25 | 2019-04-26 | 深圳市迈腾电子有限公司 | The detection device of wireless router circuit board |
WO2020155299A1 (en) * | 2019-02-01 | 2020-08-06 | 网宿科技股份有限公司 | Method for fitting target object in video frame, system, and device |
JP2022503426A (en) * | 2019-09-27 | 2022-01-12 | ベイジン センスタイム テクノロジー デベロップメント カンパニー, リミテッド | Human body detection methods, devices, computer equipment and storage media |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107305686A (en) * | 2016-04-20 | 2017-10-31 | 掌赢信息科技(上海)有限公司 | A kind of image processing method and electronic equipment |
CN107808137A (en) * | 2017-10-31 | 2018-03-16 | 广东欧珀移动通信有限公司 | Image processing method, device, electronic equipment and computer-readable recording medium |
CN107808136A (en) * | 2017-10-31 | 2018-03-16 | 广东欧珀移动通信有限公司 | Image processing method, device, readable storage medium storing program for executing and computer equipment |
-
2018
- 2018-09-17 CN CN201811085280.5A patent/CN109242868B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107305686A (en) * | 2016-04-20 | 2017-10-31 | 掌赢信息科技(上海)有限公司 | A kind of image processing method and electronic equipment |
CN107808137A (en) * | 2017-10-31 | 2018-03-16 | 广东欧珀移动通信有限公司 | Image processing method, device, electronic equipment and computer-readable recording medium |
CN107808136A (en) * | 2017-10-31 | 2018-03-16 | 广东欧珀移动通信有限公司 | Image processing method, device, readable storage medium storing program for executing and computer equipment |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109683084A (en) * | 2019-01-25 | 2019-04-26 | 深圳市迈腾电子有限公司 | The detection device of wireless router circuit board |
WO2020155299A1 (en) * | 2019-02-01 | 2020-08-06 | 网宿科技股份有限公司 | Method for fitting target object in video frame, system, and device |
JP2022503426A (en) * | 2019-09-27 | 2022-01-12 | ベイジン センスタイム テクノロジー デベロップメント カンパニー, リミテッド | Human body detection methods, devices, computer equipment and storage media |
JP7101829B2 (en) | 2019-09-27 | 2022-07-15 | ベイジン・センスタイム・テクノロジー・デベロップメント・カンパニー・リミテッド | Human body detection methods, devices, computer equipment and storage media |
Also Published As
Publication number | Publication date |
---|---|
CN109242868B (en) | 2021-05-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110826519B (en) | Face shielding detection method and device, computer equipment and storage medium | |
CN108197532B (en) | The method, apparatus and computer installation of recognition of face | |
CN109934115B (en) | Face recognition model construction method, face recognition method and electronic equipment | |
CN106295567B (en) | A kind of localization method and terminal of key point | |
US20230119593A1 (en) | Method and apparatus for training facial feature extraction model, method and apparatus for extracting facial features, device, and storage medium | |
CN106875422B (en) | Face tracking method and device | |
CN109146892A (en) | A kind of image cropping method and device based on aesthetics | |
CN108460338A (en) | Estimation method of human posture and device, electronic equipment, storage medium, program | |
CN106874826A (en) | Face key point-tracking method and device | |
CN108986075A (en) | A kind of judgment method and device of preferred image | |
CN109598234A (en) | Critical point detection method and apparatus | |
CN109242868A (en) | Image processing method, device, electronic equipment and storage medium | |
WO2019228040A1 (en) | Facial image scoring method and camera | |
CN110458235B (en) | Motion posture similarity comparison method in video | |
CN109934182A (en) | Object behavior analysis method, device, electronic equipment and computer storage medium | |
CN111401219B (en) | Palm key point detection method and device | |
CN109993021A (en) | The positive face detecting method of face, device and electronic equipment | |
CN112101124A (en) | Sitting posture detection method and device | |
CN113763348A (en) | Image quality determination method and device, electronic equipment and storage medium | |
CN107590460A (en) | Face classification method, apparatus and intelligent terminal | |
CN110175519A (en) | Method and device for identifying separation and combination identification instrument of transformer substation and storage medium | |
CN108520263A (en) | A kind of recognition methods of panoramic picture, system and computer storage media | |
CN111652168B (en) | Group detection method, device, equipment and storage medium based on artificial intelligence | |
CN111784660B (en) | Method and system for analyzing frontal face degree of face image | |
CN111553250B (en) | Accurate facial paralysis degree evaluation method and device based on face characteristic points |
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 |