1. 首页
  2. 文章中心
  3. 深度学习

Feature Transfer Learning for Deep Face Recognition with Long-Tail Data 论文阅读笔记

133 阅读 0 评论
我是靠谱客的博主 坦率荔枝，这篇文章主要介绍Feature Transfer Learning for Deep Face Recognition with Long-Tail Data 论文阅读笔记，现在分享给大家，希望可以做个参考。

 

问题： 训练人脸模型需要依赖大规模数据，比如 MsCeleb1M,deepglint, WebFace 等，但这些大规模数据存在严重的long-tail 问题。 简单说，20%人占了80%数据。   这个问题普遍存在，也有不少paper解决该问题。 比如range loss 等。 

本文idea： 将GAN 网络方法融合到主流的人脸识别模型中， 在特征领域进行迁移学习。  

本文方法比较新颖，先简单记录一下，抽时间精读时，再做详细的笔记。

 

本文

 

 

 

 

 

 

 

最后

以上就是坦率荔枝最近收集整理的关于Feature Transfer Learning for Deep Face Recognition with Long-Tail Data 论文阅读笔记的全部内容，更多相关Feature内容请搜索靠谱客的其他文章。

本图文内容来源于网友提供，作为学习参考使用，或来自网络收集整理，版权属于原作者所有。
点赞(88)

相关文章

【模型压缩】蒸馏算法小结模型压缩之蒸馏算法小结
【模型压缩】蒸馏算法小结模型压缩之蒸馏算法小结
【论文阅读】Multisource Transfer Learning With Convolutional Neural Networks for Lung Pattern AnalysisMultisource Transfer Learning With Convolutional Neural Networks for Lung Pattern Analysis
【论文阅读】Multisource Transfer Learning With Convolutional Neural Networks for Lung Pattern AnalysisMultisource Transfer Learning With Convolutional Neural Networks for Lung Pattern Analysis
深度迁移学习综述：A Survey on Deep Transfer Learning
深度迁移学习综述：A Survey on Deep Transfer Learning
Summary of defect detection algorithms based on deep learning基于深度学习的缺陷检测算法汇总1、A fast and robust convolutional neural network-based defect detection model in product quality control2、Automatic Defect Detection of Fasteners on the Catenary Support Devi
Summary of defect detection algorithms based on deep learning基于深度学习的缺陷检测算法汇总1、A fast and robust convolutional neural network-based defect detection model in product quality control2、Automatic Defect Detection of Fasteners on the Catenary Support Devi
Feature Transfer Learning for Deep Face Recognition with Long-Tail Data 论文阅读笔记
Feature Transfer Learning for Deep Face Recognition with Long-Tail Data 论文阅读笔记
吴恩达深度学习课程deeplearning.ai课程作业：Class 4 Week 4 Art Generation with Neural Style TransferDeep Learning & Art: Neural Style Transfer
吴恩达深度学习课程deeplearning.ai课程作业：Class 4 Week 4 Art Generation with Neural Style TransferDeep Learning & Art: Neural Style Transfer
深度学习实现缺陷检测1、A fast and robust convolutional neural network-based defect detection model in product quality control2、Automatic Defect Detection of Fasteners on the Catenary Support Device Using Deep Convolutional Neural Network（基于深度卷积网络的接触网支架紧固件缺陷自动检测
深度学习实现缺陷检测1、A fast and robust convolutional neural network-based defect detection model in product quality control2、Automatic Defect Detection of Fasteners on the Catenary Support Device Using Deep Convolutional Neural Network（基于深度卷积网络的接触网支架紧固件缺陷自动检测
基于特征的对抗迁移学习论文_<EYD与机器学习>迁移学习：DANN域对抗迁移网络...
基于特征的对抗迁移学习论文_<EYD与机器学习>迁移学习：DANN域对抗迁移网络...

评论列表共有 0 条评论

立即
投稿
返回
顶部