4 results about "Face wrinkling" patented technology

The invention discloses a face wrinkle detection method and system based on dual-branch feature fusion and a multi-input attention mechanism. The method comprises the following steps: dual-branch feature extraction: extracting local texture features through a ResNet branch, and modeling global context dependence through a Swin Transform branch; the feature fusion module adopts 1 * 1 convolution to align channel dimensions (ResNet branch compression ratio is 4: 1, and Swin branch compression ratio is 1.5: 1) output by double branches, and feature fusion is realized through splicing or adding; a multi-input channel attention mechanism: performing multi-scale convolution on the fusion features to generate three paths of inputs, adding the three paths of inputs, performing compression-excitation operation to generate three groups of independent channel weights, and performing weighted fusion to output enhanced features; and the classification module is used for outputting the fine-grained wrinkle category probability. The system comprises a processor and a memory and is used for implementing the method. According to the method, the wrinkle detection precision is remarkably improved, and the method is superior to a mainstream model in fine-grained classification tasks such as nasolabial folds and crow's feet, and is suitable for the fields of medical beauty evaluation and skin analysis.

The invention discloses an intelligent face wrinkle removal method and system based on a generative adversarial network. The method comprises the following steps: S1, performing face detection and face key point detection on an original first image; s2, performing face extraction and face correction on the first image to obtain a second image; s3, inputting the second image into a wrinkle segmentation model to obtain a wrinkle mask image, and adopting an improved unet network structure; s4, splicing the second image and the wrinkle mask image to obtain a third image, inputting the third image into the face wrinkle removal model to obtain a fourth image, and adopting an improved generative adversarial network structure; s5, subtracting the fourth image from the second image, and performing inverse affine transformation to obtain a difference graph; s6, adding the difference image into the first image to obtain a fifth image; and S7, protecting a non-face skin area in the fifth image to obtain a final wrinkle removal result image. According to the invention, through combination of precise segmentation and intelligent wrinkle removal technologies, efficient and natural beautifying experience is provided for the user.

The invention relates to the technical field of image processing, and discloses an occluded face wrinkle detection method, system, device and medium, and the method comprises the steps: carrying out the key point detection of an eye region in a face image through a MediaPipe face detector, roughly positioning the face region in the face image through the key point information of the eye region, and carrying out the recognition of the face region in the face image; and inputting the face area into the trained deep learning segmentation model, outputting a wrinkle segmentation mask image, and respectively carrying out differential edge detection on the face area in the horizontal direction and the vertical direction to obtain a face vertical grain image and a face transverse grain image, so that the edge features of the face wrinkles can be well detected, and the accuracy of the face wrinkle segmentation is improved. According to the method, the detection error caused by illumination change is avoided, and then the facial vertical grain image and the facial transverse grain image are superposed, so that the accuracy and robustness of facial wrinkle detection are accurately improved.

The application provides a method for removing wrinkles when generating a human face. The method includes constructing a large-scale human face image dataset; using a multi-scale edge detection method to extract edge features and fuse them into a single composite image; positioning the facial contour area based on a human face key point detection technology to generate a human face wrinkle feature map; inputting text description information and an original human face image into a human face diffusion model, and simultaneously inputting the wrinkle feature map as a conditional constraint into a ControlNet module; guiding the generation process of the diffusion model through the ControlNet module to realize an end-to-end joint training framework; generating a preliminary human face image output in an initial generation stage; constructing a comprehensive edge representation through multi-scale edge feature fusion; constructing a feature image representing wrinkle distribution through a wrinkle feature map; and generating an optimization through a double guidance mechanism to generate a beautification effect. In the finally generated picture, the human face is whiter, wrinkles are removed, and the overall aesthetic sense is obviously improved.