State-of-the-art 2D and 3D Face Analysis Project

Implementation of popular deep learning networks with TensorRT network definition API

YOLO5Face: Why Reinventing a Face Detector (https://arxiv.org/abs/2105.12931) ECCV Workshops 2022)

RetinaFace: Deep Face Detection Library for Python

基于Yolov5车牌检测,更快更准.

nniefacelib是一个在海思35xx系列芯片上运行的人脸算法库

TensorRT-7 Network Lib 包括常用目标检测、关键点检测、人脸检测、OCR等 可训练自己数据

yolov8 face detection with landmark

InsightFace REST API for easy deployment of face recognition services with TensorRT in Docker.

yolov7 face detection with landmark

The remake of the https://github.com/biubug6/Pytorch_Retinaface

适用于复杂场景的人脸识别身份认证系统

Face Analysis: Detection, Age Gender Estimation & Recognition

Reimplement RetinaFace use C++ and TensorRT

RetinaFace (Single-stage Dense Face Localisation in the Wild, 2019) implemented (ResNet50, MobileNetV2 trained on single GPU) in Tensorflow 2.0+. This is an unofficial implementation. With Colab.

基于人脸关键区域提取的人脸识别（LFW:99.82%+ CFP_FP:98.50%+ AgeDB30:98.25%+）

RetinaFace (RetinaFace: Single-stage Dense Face Localisation in the Wild, published in 2019) reimplemented in Tensorflow 2.0, with pretrained weights available !

Support Yolov5(4.0)/Yolov5(5.0)/YoloR/YoloX/Yolov4/Yolov3/CenterNet/CenterFace/RetinaFace/Classify/Unet. use darknet/libtorch/pytorch/mxnet to onnx to tensorrt

[CVPR 2020] Reimplementation of RetinaFace, faster and stronger.

Fast and reliable face detection with RetinaFace.PyTorch