cmusatyalab/openface

Face recognition with Google's FaceNet deep neural network.

Sagar

@sagarkpatil3

@xugaoxiang import error

@xugaoxiang https://stackoverflow.com/questions/50248017/importerror-no-module-named-openface-in-windows-8-1

Konstantinos Tsivis

@KonstantinosTsivis

Hello, can someone explain me how can I install a demo of the application please?

Konstantinos Tsivis

@KonstantinosTsivis

oh nvm I think I've found it ^^

Geotyper

@geotyper

Recurrent Neural Network Genetic Algorithm learning to play Agairo in multiworld https://youtu.be/O5RwsW0Bsrw

Xu GaoXiang

@xugaoxiang

how to use facenet model in openface? \

DerBAdii

@DerBAdii

hello everybody
how i can run openface in c# app

have you any idea about using API of openface in C#

Aditya Addepalli

@Dyex719

Hi everyone! I wanted to know how the pre-trained models were generated. Specifically, I want the architecture and details of the last layer using in training for generation of the pre-trained models. Thanks a bunch.

LINWANGQING

@LINWANGQING

hello ，i want to ask one question，why i use facenet to test on the LFW

hello ，i want to ask one question，why i use facenet to test on the LFW the result is terrible? ths

Zohaib

@netdevilz_twitter

Guys, how well it works? I setup web demo and it's showing people wrong even though it is trained ok and not showing any unknown. Can someone explain how accurate it is in general? Does it only detect one face at a time?

Gant Laborde

@GantMan

Greetings fellow engineers! I’m looking for sage advice on a project. I want to write a phone app that identifies a single person.

I was wondering if anyone has successfully ported this python lib to iPhone.

Yehor Sikachov

@exotikh3

@GantMan i highly do not recommend to port this library

you could you tensorflow lite as a framework for run dl models on phones and what you need is to port an architecture. you could find this at google with request "facenet"

Italo José

@italojs

some one already installed the openFace in arch linux(manjaro)?

Xu GaoXiang

@xugaoxiang

@netdevilz_twitter , openface support multi-face in an image, it works well on lcoal camera, but not well on IP camera(ONVIF), and it need more images to train that is not feasible in real world. I have seen a product that only need one picture per person and doing recognition well .The worst thing is openface do not recognize unknown person.

ArshiaKaul

@ArshiaKaul

Hello! I'm using Android Studio and need some help with the installation process and use of the API. I intend to compare an image captured by the application to all the images in a dataset and return the most similar image.

plumos

@plumos

Hello! Can anybody tell what does the model such as nn4.v1, nn4.small1.v1 mean in openface?

rodrigoruiz

@rodrigoruiz

Hi, I'm not sure this is the right room, but I'm trying to deploy openface to Heroku and I'm getting the following error:

Traceback (most recent call last):
File "./demos/web/websocket-server.py", line 22, in
import txaio
ImportError: No module named txaio
So I tried adding txaio >= 2.10.0 to the requirements.txt, but nothing changed.

Any idea what that might be?

JIA Pei

@jiapei100

Am I here?

That's great...

JaishreeJaishankar

@JaishreeJaishankar

Hello ! Can't we use openface for open set data ? That is I will use a pre trained model and then give the query images to perform the facial recognition on the given test frames.A model like facenet or sphereface.

Nareshchary

@Nareshchary

hi everybody ! Is openFace code available in python only? i know python language only in openface repo lua language also included. if anyone tried with only in python please help me out

rvpisal

@rvpisal

Hi There,

I was trying to understand the code by cowbjt on https://github.com/cowbjt/openface/tree/demo-web-stand-alone. I wanted to detect "unknown" face when the classifier is called. But I am struggling to understand where can i program it. Can anyone please help me out here.

Fraser-Paine

@Fraser-Paine

Hey all, does anyone know if the OpenFace model has ever been evaluated on the WIDERFace data set?
http://mmlab.ie.cuhk.edu.hk/projects/WIDERFace/
This seems like the new benchmark for state of the art models but I can't find any record of OpenFace being run on it

syedmahmoodepoch

@syedmahmoodepoch

Hi all. Can anyone explain how can we import openface in django or if not how to implement it on web

Tamtle

@Tamtle

Hi everybody, I'm Tam, I'm sure this question was ask several time before, but I can't find an answer to this: Is there a way to build and run Openface on Windows (7)? Thank you!

akhilgupta0221

@akhilgupta0221

Hi Can use Openface for the profile/side face recognition

phamngocquy

@phamngocquy

how can i re-train model when have new data.

soumenms2015

@soumenms2015

how can I install bamos/openface on windows 7?

Aml fawzy

@AmlFawzy

hello every body

i see post that taking about triple loss technique with open face

what i wanna say i wanna help to use this technique to build a hand recognition model like open face can any one help ?

this is the link http://bamos.github.io/2016/01/19/openface-0.2.0/

Abhijeet Chauhan

@abhijeetchauhan

Can anyone send me the link on the code from facenet where triplets are selected?

应江涛

@M5QzljXZTVZVzrk_twitter

kiemkhach

@kiemkhach

hello

kiemkhach

@kiemkhach

Hi.
I want to re-train model with new data, and I have to estimate time and GPU for training model.
If I have 1 tesla p100 gpu, how much batchsize should I set and how much time will it take to complete 1 million photos (batch 96x96)? who can help me ? Note: I have 100.000 people.

Amir Rajak

@taureanamir

Hello,

I have fine tuned the openface on my custom dataset and achieved very good verification accuracy. Now I need to use the obtained model in a realtime demo using my webcam as done in https://github.com/davidsandberg/facenet/blob/master/contributed/real_time_face_recognition.py. Can anyone provide some ideas on how can I achieve this?

Sujay S Kumar

@SujaySKumar

Hey guys. I am trying to perform face recognition on 52k people. Each person has a single mugshot. I am calculating L2 distance between the input image and all 52k images. The closest distances are in the range of 0.1-0.2 and they are incorrect. Does openface fail in scenarios where the number of people are high?