Face key point detection method and system based on sparse key point calibration
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A technology of face key points and detection methods, which is applied in the field of face detection to achieve the effects of improving detection accuracy, improving stability, and reducing dependencies
Active Publication Date: 2020-02-21
HANGZHOU QUWEI SCI & TECH
View PDF7 Cites 1 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
This brings great challenges to the current face key point detection method based on the face frame.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0044] Such as figure 1 As shown, this embodiment proposes a face key point detection method based on sparse key point calibration, including:
[0045] S1. Calculate the average face of dense key points under the input size of the detection model;
[0046] Face key point detection includes face key point detection, positioning or face alignment, which refers to locating the key areas of the face for a given face image, including eyebrows, eyes, nose, mouth, and facial contours. The present invention uses a deep learning method to detect key points, and the specific detection method is not limited here.
[0047] Before using the detection model for key point detection, it is necessary to train and generate the corresponding detection model. The production of training samples is based on the average face as a standard, and a certain proportion of disturbance enhancement is performed to ensure the consistency of the distribution of training samples and predicted samples. Gener...
Embodiment 2
[0063] Such as figure 2 As shown, this embodiment proposes a face key point detection system based on sparse key point calibration, including:
[0064] The mean face calculation module is used to calculate the mean face of dense key points under the input size of the detection model;
[0065] Face key point detection includes face key point detection, positioning or face alignment, which refers to locating the key areas of the face for a given face image, including eyebrows, eyes, nose, mouth, and facial contours. The present invention uses a deep learning method to detect key points, and the specific detection method is not limited here.
[0066] Before using the detection model for key point detection, it is necessary to train and generate the corresponding detection model. Production of training samples The mean face is used as a standard to generate a certain percentage of disturbance enhancements to ensure the consistency of the distribution of training samples and pre...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
PUM
Login to view more
Abstract
The invention discloses a face key point detection method and system based on sparse key point calibration. The detection method comprises the following steps: S1, calculating a mean value face of dense key points under the input size of a detection model; S2, detecting sparse key points of the face image by using an existing face detector; S3, calculating an affine transformation matrix based onthe mean human face and the sparse key points of the human face image; S4, performing affine transformation on the face image to an input image of the mean face size based on the affine transformationmatrix; and S5, detecting key points of the face image based on the detection model, and restoring the coordinates of the key points through inverse affine transformation to obtain dense key points in the original face image. According to the invention, the sparse key points are detected, the input image is obtained through affine transformation, the position of the face is closer to the proportion of the face, and the detection precision of the face key points is improved. Meanwhile, the dependence of key point detection on the face frame is reduced, and the stability is improved.
Description
technical field [0001] The invention relates to the field of human face detection, in particular to a human face key point detection method and system based on sparse key point calibration. Background technique [0002] With the development of deep learning technology, algorithms represented by neural networks have made unprecedented breakthroughs in many fields. Among them, as a classic problem in computer vision, face key point detection, whether it is based on deep learning or traditional methods, has a common shortcoming-over-reliance on face frames. In the existing methods, a detection algorithm often only performs better for the face frame used in training, and when another face frame is used, the detection accuracy of face key points will decrease. Even if the same standard face frame is used, the instability of the detector will cause the instability of the face frame, which will also cause great interference to the detection of key points of the face. The main perf...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more