search

GNAYUOHZ / ReID-MGN

Simple pytorch unofficial implement of paper: Learning Discriminative Features with Multiple Granularities for Person Re-Identification
235 stars 61 forks source link

A quetions about the initial distances for re-ranking. #12

Closed huge123 closed 5 years ago

huge123 commented 5 years ago 
 q_g_dist = np.dot(qf, np.transpose(gf)) 
 q_q_dist = np.dot(qf, np.transpose(qf))  
 g_g_dist = np.dot(gf, np.transpose(gf)) 
 dist = re_ranking(q_g_dist, q_q_dist, g_g_dist)

Thanks for sharing the codes, but I have a quetion when I read it. why the initial distance matrices for re-ranking are the correlation matrices of features, not the Mahalanobis distance ?

GNAYUOHZ commented 5 years ago

I think it may be ok using different distance

huge123 commented 5 years ago

Thanks, I think I have got the answer