CN112464715A - Sit-up counting method based on human body bone point detection - Google Patents
Sit-up counting method based on human body bone point detection Download PDFInfo
- Publication number
- CN112464715A CN112464715A CN202011140756.8A CN202011140756A CN112464715A CN 112464715 A CN112464715 A CN 112464715A CN 202011140756 A CN202011140756 A CN 202011140756A CN 112464715 A CN112464715 A CN 112464715A
- Authority
- CN
- China
- Prior art keywords
- sit
- human
- feature map
- point detection
- method based
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 title claims abstract description 20
- 210000000988 bone and bone Anatomy 0.000 title claims abstract description 19
- 238000012360 testing method Methods 0.000 claims abstract description 16
- 238000007781 pre-processing Methods 0.000 claims abstract description 3
- 238000002360 preparation method Methods 0.000 claims abstract description 3
- 238000012549 training Methods 0.000 claims description 16
- 230000004927 fusion Effects 0.000 claims description 14
- 230000001154 acute effect Effects 0.000 claims description 4
- 238000005452 bending Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 4
- 238000010586 diagram Methods 0.000 claims description 2
- 239000011159 matrix material Substances 0.000 claims description 2
- 238000005070 sampling Methods 0.000 claims description 2
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims 2
- 210000003423 ankle Anatomy 0.000 claims 2
- 210000003127 knee Anatomy 0.000 claims 2
- 230000009466 transformation Effects 0.000 claims 2
- 210000000707 wrist Anatomy 0.000 claims 2
- 238000000926 separation method Methods 0.000 abstract 1
- 230000006870 function Effects 0.000 description 7
- 230000004913 activation Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 3
- 238000011423 initialization method Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000013100 final test Methods 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- 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
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B71/00—Games or sports accessories not covered in groups A63B1/00 - A63B69/00
- A63B71/06—Indicating or scoring devices for games or players, or for other sports activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
- G06V20/42—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items of sport video content
-
- 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
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B2220/00—Measuring of physical parameters relating to sporting activity
- A63B2220/17—Counting, e.g. counting periodical movements, revolutions or cycles, or including further data processing to determine distances or speed
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Evolutionary Biology (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Physical Education & Sports Medicine (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a sit-up counting method based on human body bone point detection, which comprises the following steps: s1, acquiring a real-time video stream of the tested person in sit-up; s2, performing frame separation on the real-time video stream to store the real-time video stream into an image, and performing preprocessing operation on the stored image; s3, sending the preprocessed image into a human skeleton point detection network for human skeleton point detection; s4, judging whether the tested person is in a preparation state or not through the human skeleton points; s5, when in the ready state, judging whether to complete a complete sit-up through the human body posterior bone point, and when completing a complete sit-up, adding 1 to the counter. The invention can realize self-service sit-up test and counting, does not need manual interference in the test process, and improves the test efficiency.
Description
Technical Field
The invention relates to the field of computer vision and the technical field of sports equipment, in particular to a sit-up counting method based on human body bone point detection.
Background
The sit-up is an important sport in domestic student physical training and military physical training, and most of sit-up tests at present are that testers lie on a cushion for testing, and then statistics people count. Exercising and testing in this manner requires more human resources and if the actions are not standardized, the test results may not meet the accuracy requirements of the final test.
Still another type of counting implements motion specification by wearing the electronic device on the tester, which makes the tester feel tethered and experience poor.
Therefore, the test method provided by the invention does not have good test experience and counting precision, and has low test efficiency and poor test experience.
Disclosure of Invention
The invention aims to provide a sit-up counting method based on human body bone point detection.
The technical solution for realizing the purpose of the invention is as follows: a sit-up counting method based on human body bone point detection comprises the following steps:
s1, acquiring a real-time video stream of the tested person in sit-up;
s2, performing frame-by-frame storage on the real-time video stream to generate an image, and performing preprocessing operation on the stored image;
s3, sending the preprocessed image into a human skeleton point detection network for human skeleton point detection to obtain a human skeleton point bitmap;
s4, judging whether the tested person is in a preparation state or not through the human skeleton points;
s5, when in the ready state, judging whether to finish a sit-up through the human body posterior bone point, and when finishing a complete sit-up, adding 1 to the counter.
Compared with the prior art, the invention has the following remarkable advantages: 1) the invention realizes the sit-up counting through deep learning, so that the sit-up counting is more accurate and the precision is higher. 2) The invention does not need to wear any electronic equipment for sit-up testers on site, and the testers experience better during testing.
Drawings
Fig. 1 is a flowchart of a sit-up counting method based on human skeletal point detection according to the present invention.
Fig. 2 is a body posture point diagram.
Detailed Description
The invention is further described with reference to the drawings and examples.
With reference to fig. 1, the invention relates to a sit-up counting method based on bone point detection, which comprises the following steps:
step one, the human body skeleton point detection network structure comprises a backbone network, a feature fusion network and a detection network. Its backbone network inherits the structure of the darknet53 network. The feature fusion network performs feature fusion on the third downsampling feature map A, the fourth downsampling feature map B and the fifth downsampling feature map C. The sizes of the third downsampled feature map a, the fourth downsampled feature map B and the fifth downsampled feature map C are 52 × 52 × 128,26 × 26 × 256 and 13 × 13 × 512 respectively. And performing 1 × 1 convolution on the fifth downsampling feature map C to change the number of channels to 256, performing upsampling and fourth downsampling feature map fusion on the fifth downsampling feature map C to form a new fourth downsampling feature map D, performing 1 × 1 convolution on the new fourth downsampling feature map D to change the number of channels to 128, and performing upsampling and third downsampling feature map A fusion on the new fourth downsampling feature map D to form a new third downsampling feature map E.
As regards the activation function, the invention uses the leak-ReLu as an activation function, since the leak-ReLu activation function has advantages in this province including: when the functions such as Sigmoid are adopted, the calculation amount is huge when the activation function is calculated, the calculation amount required by derivation is too large when the error gradient is calculated by back propagation, and the whole calculation amount is saved by adopting the Leaky-ReLu activation function. For a deep network, when the Sigmoid function is reversely propagated, the phenomena of gradient explosion and gradient dispersion can easily occur, and the Leaky-ReLu can effectively solve the problem.
Regarding the selection of the feature fusion layer, the feature fusion is performed to obtain the final third-time downsampling feature map E, the fourth-time downsampling feature map D and the fifth-time downsampling feature map C, wherein the feature map sizes of the final third-time downsampling feature map E, the fourth-time downsampling feature map D and the fifth-time downsampling feature map C are 52 × 52 × 128,26 × 26 × 256 and 13 × 13 × 512 respectively. And 5 times of 3 × 3 convolution operations are performed on each feature map to form new feature maps F, G and H, respectively. And passing F, G, H through the output layer, respectively.
Regarding the selection of the output layer, the output layer of the present invention is a matrix of W/s × H/s × C, where W represents the width of the image in the input network, H represents the height of the image in the input network, and C represents the classification number of the last output keypoint. S represents a multiple of the down-sampling.
Step two, training the model:
firstly, normalization operation is carried out on a selected sample human body image, each person is cut out, and human body bone points are labeled.
Secondly, the image is stretched in multiple scales by means of translation invariance of data, the stretching is 1.1 times or 1.2 times of the original stretching in multiple scales, the image is rotated by an angle of-15 degrees to 15 degrees, and the image is turned over and processed in a mirror image mode to obtain more training images in different scales.
Thirdly, dividing the whole data set into K parts, selecting one subset as a test set each time, selecting 80 percent of the K-1 subsets as a training set and 20 percent of the K-1 subsets as a verification set, and performing K times of cross verification, thereby training the network model.
Fourthly, when the neural network model is trained, the convolution part uses the pre-training weight of the darknet53, and a random initialization method is adopted in the feature fusion layer, so that the training time can be reduced, and a better detection effect can be obtained in less time. In addition, some hyper-parameters are required to be set, including the number of iterations, the size of the image quantity batch-size input to the neural network for training each time is set, and the training end condition is determined. In the present invention, the value of epoch is set to 50 and the value of batch-size is set to 64. By setting initial network weight and adopting a random initialization method for network parameters, iterative training is continuously carried out until loss is less than a set threshold value or the iteration times is greater than the set threshold value, and the training is finished.
Fifth, the method comprises the following steps: obtaining a trained neural network model
Thirdly, shooting a video of the human body posture test area by the camera, and reading an image frame of the video shot by the camera in real time; preferably, the camera is separately installed above the test area, and the image of the camera covers the position of the test area to be read;
acquiring a detection image, and carrying out normalization operation on the resolution ratio of the detection image;
and sending the normalized picture into a trained human body posture model for human body skeleton point detection, positioning coordinates of 18 skeleton points in a human body through the human body posture model, scoring each skeleton point, and displaying the detected skeleton point in an image to be detected. With reference to fig. 2, the region formed by the human skeleton points 2, 3, 4 is an acute triangle, and the region formed by the human skeleton points 5, 6, 7 is an acute triangle, so that the tester is ready to send out, the maximum distance between the human skeleton point 0 and the human skeleton point 10 at the moment is recorded and recorded as a threshold value a, and the half of the maximum distance between the human skeleton point 0 and the human skeleton point 13 at the moment is recorded and recorded as a threshold value B; when the maximum distance between the human skeleton point 0 and the human skeleton points 10 and 13 is smaller than a threshold value B, a bending signal is sent out; when the maximum distance between the human skeleton point 0 and the human skeleton points 10 and 13 is greater than or equal to the threshold value A, a straightening state is recorded, and when a continuous bending and straightening state signal appears, the sit-up counter is increased by 1.
Claims (10)
1. A sit-up counting method based on human body bone point detection is characterized by comprising the following steps:
s1, acquiring a real-time video stream of the tested person in sit-up;
s2, performing frame-by-frame storage on the real-time video stream to generate an image, and performing preprocessing operation on the stored image;
s3, sending the preprocessed image into a human skeleton point detection network for human skeleton point detection to obtain a human skeleton point bitmap;
s4, judging whether the tested person is in a preparation state or not through the human skeleton points;
s5, when in the ready state, judging whether to finish a sit-up through the human body posterior bone point, and when finishing a complete sit-up, adding 1 to the counter.
2. The sit-up method based on human skeletal point detection as claimed in claim 1, wherein: in step s1, the real-time video stream of the tested person is shot, and the real-time video stream image is analyzed to obtain the sit-up information of the tested person.
3. The sit-up method based on human skeletal point detection as claimed in claim 1, wherein: in step s2, the analyzed image is corrected to a rectangular shape, and the image resolution is normalized.
4. The sit-up method based on human skeletal point detection as claimed in claim 1, wherein: in step s3, the human body skeleton point detection network structure comprises a backbone network, a feature fusion network and a detection network; the backbone network inherits the darknet53 network structure; the feature fusion network performs feature fusion on the third downsampling feature map A, the fourth downsampling feature map B and the fifth downsampling feature map C; the sizes of the third downsampling feature map A, the fourth downsampling feature map B and the fifth downsampling feature map C are respectively 52 × 52 × 128,26 × 26 × 256 and 13 × 13 × 512; and performing 1 × 1 convolution on the fifth downsampling feature map C to change the number of channels to 256, performing upsampling and fourth downsampling feature map fusion on the fifth downsampling feature map C to form a new fourth downsampling feature map D, performing 1 × 1 convolution on the new fourth downsampling feature map D to change the number of channels to 128, and performing upsampling and third downsampling feature map A fusion on the new fourth downsampling feature map D to form a new third downsampling feature map E.
5. The sit-up counting method based on human skeletal point detection as claimed in claim 4, wherein: the feature map sizes of the final third downsampling feature map E, the fourth downsampling feature map D and the fifth downsampling feature map C obtained through feature fusion are respectively 52 × 52 × 128,26 × 26 × 256 and 13 × 13 × 512; and 5 times of 3 × 3 convolution operations are performed on each feature map to form new feature maps F, G and H respectively, and the new feature maps F, G and H are passed through the output layer respectively.
6. The sit-up counting method based on human skeletal point detection as claimed in claim 5, wherein: the output layer is a matrix of W/s × H/s × C, wherein W represents the width of the image in the input network, H represents the height of the image in the input network, C represents the classification number of the final output key point, and s represents a multiple of down-sampling.
7. The sit-up method based on human skeletal key point detection as claimed in claim 1, wherein: training a human skeleton point detection network model: selecting a sit-up video, cutting the sit-up video into images, calibrating a test area for sit-up in the images, carrying out affine transformation on the calibrated area, and carrying out human body key point marking on the area subjected to affine transformation to obtain a training sample; and (4) sending the training samples into a human skeleton key point detection model for training, thereby obtaining a trained human posture key point model.
8. The sit-up counting method based on human skeletal point detection as claimed in claim 1, wherein: the skeleton point diagram of the human body totally records 18 point positions, 0 represents a nose, 1 represents a neck, 2 represents a right shoulder, 3 represents a right elbow, 4 represents a right wrist, 5 represents a left shoulder, 6 represents a left elbow, 7 represents a left wrist, 8 represents a right hip, 9 represents a right knee, 10 represents a right ankle, 11 represents a left hip, 12 represents a left knee, 13 a left ankle, 14 represents a right eye, 15 represents a left eye, 16 represents a right ear, and 17 represents a left ear.
9. The sit-up counting method based on human bone point detection according to claim 1 or 8, wherein in step s4, the region formed by human bone points 2, 3 and 4 is an acute triangle, and the region formed by human bone points 5, 6 and 7 is an acute triangle, so that the tester is ready to record the maximum distance between human bone point 0 and human bone point 10 at this time as threshold a, and record the half of the maximum distance between human bone point 0 and human bone point 13 at this time as threshold B.
10. The sit-up counting method based on human skeletal point detection according to claim 1 or 9, wherein: in step s5, when it is detected that the maximum distance between the human skeleton point 0 and the human skeleton points 10 and 13 is smaller than the threshold value B, a bending signal is sent out; when the maximum distance between the human skeleton point 0 and the human skeleton points 10 and 13 is greater than or equal to the threshold value A, a straightening state is recorded, and when a continuous bending and straightening state signal appears, the sit-up counter is increased by 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011140756.8A CN112464715A (en) | 2020-10-22 | 2020-10-22 | Sit-up counting method based on human body bone point detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011140756.8A CN112464715A (en) | 2020-10-22 | 2020-10-22 | Sit-up counting method based on human body bone point detection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112464715A true CN112464715A (en) | 2021-03-09 |
Family
ID=74834147
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011140756.8A Pending CN112464715A (en) | 2020-10-22 | 2020-10-22 | Sit-up counting method based on human body bone point detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112464715A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113033515A (en) * | 2021-05-24 | 2021-06-25 | 北京每日优鲜电子商务有限公司 | Wearing detection method and device, electronic equipment and computer-readable storage medium |
| CN113487635A (en) * | 2021-07-01 | 2021-10-08 | 盛视科技股份有限公司 | Sit-up counting method based on image difference |
| CN113893515A (en) * | 2021-10-13 | 2022-01-07 | 恒鸿达科技有限公司 | Sit-up test counting method, sit-up test counting device and sit-up test counting medium based on vision technology |
| CN114259721A (en) * | 2022-01-13 | 2022-04-01 | 王东华 | Training evaluation system and method based on Beidou positioning |
| CN115171208A (en) * | 2022-05-31 | 2022-10-11 | 中科海微(北京)科技有限公司 | Sit-up posture evaluation method and device, electronic equipment and storage medium |
| CN117095152A (en) * | 2023-10-17 | 2023-11-21 | 南京佳普科技有限公司 | Bone recognition camera for physical training evaluation and training evaluation method |
| CN113487635B (en) * | 2021-07-01 | 2024-05-28 | 盛视科技股份有限公司 | Sit-up counting method based on image difference |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017080202A (en) * | 2015-10-29 | 2017-05-18 | キヤノンマーケティングジャパン株式会社 | Information processing device, information processing method and program |
| CN111368810A (en) * | 2020-05-26 | 2020-07-03 | 西南交通大学 | Sit-up detection system and method based on human body and skeleton key point identification |
| CN111401260A (en) * | 2020-03-18 | 2020-07-10 | 南通大学 | Sit-up test counting method and system based on Quick-OpenPose model |
-
2020
- 2020-10-22 CN CN202011140756.8A patent/CN112464715A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017080202A (en) * | 2015-10-29 | 2017-05-18 | キヤノンマーケティングジャパン株式会社 | Information processing device, information processing method and program |
| CN111401260A (en) * | 2020-03-18 | 2020-07-10 | 南通大学 | Sit-up test counting method and system based on Quick-OpenPose model |
| CN111368810A (en) * | 2020-05-26 | 2020-07-03 | 西南交通大学 | Sit-up detection system and method based on human body and skeleton key point identification |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113033515A (en) * | 2021-05-24 | 2021-06-25 | 北京每日优鲜电子商务有限公司 | Wearing detection method and device, electronic equipment and computer-readable storage medium |
| CN113487635A (en) * | 2021-07-01 | 2021-10-08 | 盛视科技股份有限公司 | Sit-up counting method based on image difference |
| CN113487635B (en) * | 2021-07-01 | 2024-05-28 | 盛视科技股份有限公司 | Sit-up counting method based on image difference |
| CN113893515A (en) * | 2021-10-13 | 2022-01-07 | 恒鸿达科技有限公司 | Sit-up test counting method, sit-up test counting device and sit-up test counting medium based on vision technology |
| CN113893515B (en) * | 2021-10-13 | 2022-12-27 | 恒鸿达科技有限公司 | Sit-up test counting method, sit-up test counting device and sit-up test counting medium based on vision technology |
| CN114259721A (en) * | 2022-01-13 | 2022-04-01 | 王东华 | Training evaluation system and method based on Beidou positioning |
| CN115171208A (en) * | 2022-05-31 | 2022-10-11 | 中科海微(北京)科技有限公司 | Sit-up posture evaluation method and device, electronic equipment and storage medium |
| CN117095152A (en) * | 2023-10-17 | 2023-11-21 | 南京佳普科技有限公司 | Bone recognition camera for physical training evaluation and training evaluation method |
| CN117095152B (en) * | 2023-10-17 | 2024-01-26 | 南京佳普科技有限公司 | Bone recognition camera for physical training evaluation and training evaluation method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112464715A (en) | Sit-up counting method based on human body bone point detection | |
| CN110555434B (en) | Method for detecting visual saliency of three-dimensional image through local contrast and global guidance | |
| CN106920224B (en) | A method of assessment stitching image clarity | |
| CN110852383B (en) | Target detection method and device based on attention mechanism deep learning network | |
| CN107463920A (en) | A kind of face identification method for eliminating partial occlusion thing and influenceing | |
| CN110879982B (en) | Crowd counting system and method | |
| CN104573731A (en) | Rapid target detection method based on convolutional neural network | |
| CN110619638A (en) | Multi-mode fusion significance detection method based on convolution block attention module | |
| Jiang et al. | A deep evaluator for image retargeting quality by geometrical and contextual interaction | |
| CN113034495B (en) | Spine image segmentation method, medium and electronic device | |
| CN107292299B (en) | Side face recognition methods based on kernel specification correlation analysis | |
| CN113762133A (en) | Self-weight fitness auxiliary coaching system, method and terminal based on human body posture recognition | |
| CN105184266B (en) | A kind of finger venous image recognition methods | |
| CN109461177B (en) | Monocular image depth prediction method based on neural network | |
| CN113610046B (en) | Behavior recognition method based on depth video linkage characteristics | |
| CN113947589A (en) | Missile-borne image deblurring method based on countermeasure generation network | |
| CN114241422A (en) | Student classroom behavior detection method based on ESRGAN and improved YOLOv5s | |
| CN110543916A (en) | Method and system for classifying missing multi-view data | |
| CN110705566A (en) | Multi-mode fusion significance detection method based on spatial pyramid pool | |
| CN107895145A (en) | Method based on convolutional neural networks combination super-Gaussian denoising estimation finger stress | |
| CN115138059A (en) | Pull-up standard counting method, pull-up standard counting system and storage medium of pull-up standard counting system | |
| CN115205547A (en) | Target image detection method and device, electronic equipment and storage medium | |
| CN113569805A (en) | Action recognition method and device, electronic equipment and storage medium | |
| CN115731597A (en) | Automatic segmentation and restoration management platform and method for mask image of face mask | |
| Chandaliya et al. | Conditional perceptual adversarial variational autoencoder for age progression and regression on child face |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210309 |
|
| RJ01 | Rejection of invention patent application after publication |