jaloveapple / SCSP

source code for paper Similarity Learning with Spatial Constraints for Person Re-identification

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

SCSP

source code for paper Similarity Learning with Spatial Constraints for Person Re-identification

There are two main folds.

1, The code in spatial_feat_extraction is to extract local patch descriptors and reduce the feature according to global region and local regions. please unzip the viper dataset in dataset file, and run main.m

The generated feature can be downloaded from https://yunpan.cn/OcBp84udrQUGeR (3a3e)

2, The code in spatial_learning_polynomial_map is about the similarity learning on polynomial feature map. Please put the generated feature in 1 to cache/viper/pca_new

About

source code for paper Similarity Learning with Spatial Constraints for Person Re-identification


Languages

Language:MATLAB 96.8%Language:C++ 3.1%Language:M 0.1%