thnkim / openfacepytorch Goto Github PK
View Code? Open in Web Editor NEWPyTorch module to use OpenFace's nn4.small2.v1.t7 model
License: Apache License 2.0
PyTorch module to use OpenFace's nn4.small2.v1.t7 model
License: Apache License 2.0
Hello, thnkim!
The code of openface for pytorch is help me. I am trying to run it but didn't work at last step, it's the problem of the file of 'openface_server.lua'.I'm not family with Torch and appreciate you can transform this file to pytorch !
I see that f is a tuple, consisting of (x, x_736), the result of the forward() method in the netOpenFace class. I am getting AttributeError: 'tuple' object has no attribute 'size' when I run your code as is.
Is this expected?
The following code is from the last few lines in loadOpenFace.py.
for i in range(f.size(0) - 1):
for j in range(i + 1, f.size(0)):
df = f[i] - f[j]
print(img_paths[i].split('/')[-1], img_paths[j].split('/')[-1], torch.dot(df, df))
thank you for the repository!!
I would like to mention there may be a mistake line 295 of loadOpenface.py, I believe it's f_736
almost nothing ;)
Hi, thnkim!
I'm using your codes to fine tune our face recognition network in pytorch, it works well when using single GPU (gpuDevice=0), I want to know how to run it with multiple GPU?
thanks!
Could you please tell me How to convert 'nn4.small2.v1.t7' to an HDF5 file using 'Torch - HDF5'? Thanks.
thank you very much
Can I run this on a CPU?
When I'm trying outputs_128, outputs_726 = face_net(imgs_tensor) (imgs_tensor is [3,3,96,96]) , I got an error :
ValueError: Expected more than 1 value per channel when training, got input size [1, 96, 1, 1].
can you solve that?
Thanks for the code. I have learned a lot.
But when I run the example I got that lennon with glasses was closer to clapton with glasses than to lennon without glasses.
I think this is completely wrong.
I get the same exact resulto with both: openface_20180119.pth and openface.pth
Thanks.
Hey, I'm new to face detection. It seems the codes replace the network part of the torch version openface.
But the face box detection and face align API in openface is not contained. Is there easy way to do that without install the torch version openface?
As the subject says: you provide openface.pth
and openface_20180119.pth
where the first one is referenced in your source code while the second one is newer.
Which model is the recommended one and yields best results?
Thanks in advance.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.