Improving CNN-based Person Re-identification using score Normalization

no code implementations • 1 Jul 2023 • Ammar Chouchane, Abdelmalik Ouamane, Yassine Himeur, Wathiq Mansoor, Shadi Atalla, Afaf Benzaibak, Chahrazed Boudellal

For example, without normalization, the rank-20 rate accuracies of the GRID, CUHK01, VIPeR and PRID450S datasets were 61. 92%, 83. 90%, 92. 03%, 96. 22%; however, after score normalization, they have increased to 64. 64%, 89. 30%, 92. 78%, and 98. 76%, respectively.

Metric Learning Person Re-Identification