0、人脸识别
https://github.com/google/mediapipe
人脸地标检测指南 |媒体管道 |面向开发者的 Google
1、图像换底
https://github.com/ZHKKKe/MODNet
2、图像增强
https://github.com/ZHKKKe/Harmonizer
3、 使用 Mediapipe Face 网格监视视角
https://github.com/itsvaibhav01/3D-Viewing-Angle-from-Face-Mesh
头部姿态估计原理及可视化_头部姿态估计landmark-CSDN博客
Head Pose Estimation with MediaPipe and OpenCV in Javascript | by Susanne Thierfelder | Medium
shenasa-ai/head-pose-estimation: Head Pose Estimation using Mediapipe face mesh (github.com) 优秀，姿态监测
使用 MediaPipe 和 OpenCV 的实时头部姿态估计 FaceMesh：综合指南 |由 Jaykumaran R |中等 (medium.com)
顔向き推定について調査中 #DeepLearning - Qiita
头部姿态估计原理及可视化_头部姿态估计
4、照片提亮增加
https://github.com/hermosayhl/DeepUPE_pytorch
https://github.com/dvlab-research/DeepUPE
5、去除阴影
https://github.com/tsingqguo/exposure-fusion-shadow-removal
6、照片质量检测
https://github.com/idealo/image-quality-assessment
超丰/IQA-PyTorch：👁️ 🖼️ 🔥用于图像质量评估的PyTorch工具箱，包括LPIPS，FID，NIQE，NRQM（马），MUSIQ，NIMA，DBCNN，WaDIQaM，BRISQUE，PI等... (github.com)
image-quality-assessment · GitHub Topics
https://github.com/Tencent/CenseoQoE 【腾讯】
7、Google图像分割人脸检测
https://mediapipe-studio.webapps.google.com/home
8、分割迁移
https://github.com/ZHKKKe/NeuralPreset
9、自动补光
https://zhuanlan.zhihu.com/p/199960003
https://github.com/YingqingHe/Shadow-Removal-via-Generative-Priors
10、去除眼睛及阴影
https://github.com/StoryMY/take-off-eyeglasses
https://github.com/SleyAI/Portrait-Shadow-Removal-and-Relighting-with-GANs
https://github.com/tsingqguo/exposure-fusion-shadow-removal
11、去除高光
https://jnjaby.github.io/projects/UDC/
https://github.com/hhqweasd/G2R-ShadowNet
https://github.com/jnjaby/DISCNet
12、照片融合
https://github.com/julia0607/Wide-Range-Image-Blending
13、图像超清
https://github.com/yangxy/GPEN
GitHub - TencentARC/GFPGAN: GFPGAN aims at developing Practical Algorithms for Real-world Face Restoration.
