Learning a Discriminative Null Space for Person Re-identification CVPR 2016
Posted candy
tags:
篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了Learning a Discriminative Null Space for Person Re-identification CVPR 2016相关的知识，希望对你有一定的参考价值。
运用KNFST进行降维，用于行人再识别。
原理部分 NFST KNFST
实验部分
数据库
VIPeR: 论文提供了提取特征后的数据。
VIPeR contains 632 identities and each has two images captured outdoor from two views with distinct view angles. All images are scaled to 128 × 48 pixels。
632 people’s images are randomly divided into two equal halves, one for training and the other for testing. This is repeated for 10 times and the averaged performance is reported. A和B两个场景中各选择一半作为训练集合，另外一半作为测试集合。
A，B的训练集合集成在一起形成训练集。A的测试作为gallary, B的测试作为probe. 最终通过训练得到对特征空间的投影矩阵（隐式的，核方法），计算样本在低维空间中的维度，然后利用欧式距离进行识别，评估采用累积匹配曲线， CMC Cumulated Matching Characteristics。
以上是关于Learning a Discriminative Null Space for Person Re-identification CVPR 2016的主要内容，如果未能解决你的问题，请参考以下文章
Center Loss - A Discriminative Feature Learning Approach for Deep Face Recognition
理解一下generative learning and discriminative learning algorithm
Learning Deep Features for Discriminative Localization
[翻译]Learning Deep Features for Discriminative Localization
DiMP: Learning Discriminative Model Prediction for Tracking
Learning Discriminative and Transformation Covariant Local Feature Detectors实验环境搭建详细过程