TF-siamesenet

KO | EN

gitlite — search

#javascript #typescript #llm #go #android #machine-learning #rust #deep-learning #linux #windows #kubernetes #awesome-list #security #open-source

(★ 91)

Face recognition based on SiameseNet

TF-siamesenet 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# TF-siamesenet

The main work of the article is to use the SiameseNet model to achieve the function of face recognition.
The SiameseNet convolutional neural network model is as follows (detailed structure moves to GitHub, a simplified version of VGG):
![SiameseNet model](https://upload-images.jianshu.io/upload_images/4019913-5b6b5dcdcb14cba6.png?imageMogr2/auto-orient/strip%7CimageView2/2/w/1240)
![Input and output examples](https://upload-images.jianshu.io/upload_images/4019913-0e9464aa9f260866.png?imageMogr2/auto-orient/strip%7CimageView2/2/w/1240)

The data set uses Microsoft's MS-Celeb-1M public face data set. We directly download the aligned data set. First, there is a cleaned list on the network. The file name is: MS-Celeb-1M_clean_list.txt, about 160M. Then use the combination algorithm to generate two files, positive_pairs_path.txt and negative_pairs_path.txt, each about 1.5million pairs.
![Generate file example](https://upload-images.jianshu.io/upload_images/4019913-c892e8c84203fb23.png?imageMogr2/auto-orient/strip%7CimageView2/2/w/1240)

The key question is, how to define the loss function, the loss function is Logistic Regression loss function actually, it is worth noting that the last layer of the activation function should use Sigmoid Function. The Loss function is as follows:
![Loss function](https://upload-images.jianshu.io/upload_images/4019913-9061f2a26a078c24.png?imageMogr2/auto-orient/strip%7CimageView2/2/w/1240)

The next step is training, training on TITAN X for about 3 days, we can see this Loss curve:
![Loss curve](https://upload-images.jianshu.io/upload_images/4019913-de14db779ade283d.png?imageMogr2/auto-orient/strip%7CimageView2/2/w/1240)

We can see that Loss converges very well.

What about the correct rate of our model?
Haha, not very good, the correct rate on the LFW is 94%+, and the model needs to be optimized. If you have ideas to solve this problem, you can give me a message or Email me (hfrommane@qq.com).

Project source code:
https://github.com/hfrommane/TF-siamesenet

Chinese readme version is:
https://www.jianshu.com/p/1df484d9eba9

If you like, give me a star, thanks.

TF-siamesenet

TF-siamesenet 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

TF-siamesenet

TF-siamesenet 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README