AlfredXiangWu/LightCNN
A PyTorch implementation of a deep convolutional neural network for face recognition trained on noisy labeled datasets.
Not currently ranked — collecting fresh signals.
star history
This repository provides a PyTorch implementation of the Light CNN architecture for deep face representation and verification. It includes training code for CNN models (LightCNN-9, 29, and v4 variants) that achieve high accuracy on face benchmarks like LFW and MegaFace. The implementation handles noisy labels in training data and provides pretrained model weights for face verification tasks.
Frequently asked
- What is AlfredXiangWu/LightCNN?
- A PyTorch implementation of a deep convolutional neural network for face recognition trained on noisy labeled datasets.
- Is LightCNN open source?
- Yes — AlfredXiangWu/LightCNN is open source, released under the MIT license.
- What language is LightCNN written in?
- AlfredXiangWu/LightCNN is primarily written in Python.
- How popular is LightCNN?
- AlfredXiangWu/LightCNN has 962 stars on GitHub.
- Where can I find LightCNN?
- AlfredXiangWu/LightCNN is on GitHub at https://github.com/AlfredXiangWu/LightCNN.