All Projects → jdeng → Goface

jdeng / Goface

Licence: mit
Face Detector based on MTCNN, tensorflow and golang

Programming Languages

go
31211 projects - #10 most used programming language
golang
3204 projects

Projects that are alternatives of or similar to Goface

mtcnn-pytorch
pytorch implementation of face detection algorithm MTCNN
Stars: ✭ 61 (-44.04%)
Mutual labels:  mtcnn
Tensorrt demos
TensorRT YOLOv4, YOLOv3, SSD, MTCNN, and GoogLeNet
Stars: ✭ 790 (+624.77%)
Mutual labels:  mtcnn
Ncnn Mtcnn
It is a mtcnn project based on ncnn
Stars: ✭ 91 (-16.51%)
Mutual labels:  mtcnn
Face Everthing
face detection alignment recognition reconstruction ...
Stars: ✭ 257 (+135.78%)
Mutual labels:  mtcnn
Mtcnn ncnn
基于ncnn框架搭建win及android端的MTCNN人脸检测工程
Stars: ✭ 471 (+332.11%)
Mutual labels:  mtcnn
Opencv Mtcnn
An implementation of MTCNN Face detector using OpenCV's DNN module
Stars: ✭ 59 (-45.87%)
Mutual labels:  mtcnn
facenet mtcnn tensorflow inference engine
This is a modified and simplified inference engine of davidsandberg's facenet
Stars: ✭ 41 (-62.39%)
Mutual labels:  mtcnn
Awesome Face Detection
Compare with various detectors - s3fd, dlib, ocv, ocv-dnn, mtcnn-pytorch, face_recognition
Stars: ✭ 106 (-2.75%)
Mutual labels:  mtcnn
Mtcnn Caffe
Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Neural Networks
Stars: ✭ 499 (+357.8%)
Mutual labels:  mtcnn
Face Recognition Cpp
Real Time Face Recognition Detector. Over 30FPS on CPU!
Stars: ✭ 68 (-37.61%)
Mutual labels:  mtcnn
Tensorflow Mtcnn
人脸检测MTCNN算法,采用tensorflow框架编写,从理解到训练,中文注释完全,含测试和训练,支持摄像头
Stars: ✭ 302 (+177.06%)
Mutual labels:  mtcnn
Face Track Detect Extract
💎 Detect , track and extract the optimal face in multi-target faces (exclude side face and select the optimal face).
Stars: ✭ 434 (+298.17%)
Mutual labels:  mtcnn
Face Recognition
A framework for creating and using a Face Recognition system.
Stars: ✭ 60 (-44.95%)
Mutual labels:  mtcnn
Facedetection
C++ project to implement MTCNN, a perfect face detect algorithm, on different DL frameworks. The most popular frameworks: caffe/mxnet/tensorflow, are all suppported now
Stars: ✭ 255 (+133.94%)
Mutual labels:  mtcnn
Facedetector
A re-implementation of mtcnn. Joint training, tutorial and deployment together.
Stars: ✭ 99 (-9.17%)
Mutual labels:  mtcnn
MTCNN TRAIN
MTCNN_Training Scripts For Face Detection with PyTorch 0.4.0
Stars: ✭ 56 (-48.62%)
Mutual labels:  mtcnn
Dface
Deep learning face detection and recognition, implemented by pytorch. (pytorch实现的人脸检测和人脸识别)
Stars: ✭ 994 (+811.93%)
Mutual labels:  mtcnn
Tensorflow Mtcnn
C++ and python Inference only for MTCNN face detector on Tensorflow. Based on davidsandberg's facenet project:
Stars: ✭ 106 (-2.75%)
Mutual labels:  mtcnn
Joint Face Detection And Alignment
Caffe and Python implementation of Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Networks
Stars: ✭ 102 (-6.42%)
Mutual labels:  mtcnn
Mtcnn
face detection and alignment with mtcnn
Stars: ✭ 66 (-39.45%)
Mutual labels:  mtcnn

goface

Face detector/embeddings based on MTCNN, tensorflow and golang

Implementation based on https://github.com/davidsandberg/facenet . Tensorflow (1.4.1) and the golang binding are required.

Model file cmd/mtcnn.pb is converted from facenet too (see scripts/convert.py. You will need to add facenet/src to PYTHONPATH to use it). You may need to regenerate the model file for a different version of tensorflow.

The facenet protobuf model file is available for download (see instructions from facenet).

Usage

	// detection
	bs, err := ioutil.ReadFile(*imgFile)
	img, err := goface.TensorFromJpeg(bs)
	det, err := goface.NewMtcnnDetector("mtcnn.pb")
	bbox, err := det.DetectFaces(img) //[][]float32, i.e., [x1,y1,x2,y2],...

	// embeddings
	mean, std := goface.MeanStd(img)
	wimg, err := goface.PrewhitenImage(img, mean, std)
	fn, err := goface.NewFacenet("facenet.pb")
	emb, err := fn.Embedding(wimg)

See cmd/detect.go. Use go build to build the binary and run with --help.

Notes

  • Not exactly the same (e.g., nms/padding is depending on tensorflow implementation).
  • Not fully tested. Performance could a little bit worse.
  • Face landmark support not implemented.
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].