Abstract | ||
---|---|---|
Appearance-based person re-identification is very challenging, i.a. due to changing illumination, image distortion, and differences in viewpoint. Therefore, it is crucial to learn an expressive feature embedding that compensates for changing environmental conditions. There are many loss functions available to achieve this goal. However, it is hard to judge which one is the best. In related work, the experiments are only performed on the same datasets, but the use of different setups and different training techniques compromises the comparability. Therefore, we compare the most widely used and most promising loss functions under identical conditions on three different setups. We provide insights into why some of the loss functions work better than others and what additional benefits they provide. We further propose sequential training as an additional training trick that improves the performance of most loss functions. In our conclusion, we provide guidance for future usage an d research regarding loss functions for appearance-based person re-identification. Source code is available (Source code: https://www.tu-ilmenau.de/neurob/data-sets-code/re-id-loss/). |
Year | DOI | Venue |
---|---|---|
2021 | 10.1007/978-3-030-86383-8_3 | ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING, ICANN 2021, PT V |
Keywords | DocType | Volume |
Person re-identification, Deep learning, Representation learning, Loss functions, Softmax loss, Triplet hard loss, Ring loss, Center loss, Additive angular margin loss, Circle loss | Conference | 12895 |
ISSN | Citations | PageRank |
0302-9743 | 0 | 0.34 |
References | Authors | |
0 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Dustin Aganian | 1 | 0 | 0.68 |
Markus Eisenbach | 2 | 37 | 6.76 |
Joachim Wagner | 3 | 0 | 0.34 |
Daniel Seichter | 4 | 0 | 2.70 |
Horst-Michael Gross | 5 | 761 | 92.05 |