Building Computationally Efficient and Well-Generalizing Person Re-Identification Models with Metric Learning
Vladislav Sovrasov, Dmitry Sidnev · 2021
This work considers the problem of domain shift in person re-identification. Being trained on one dataset, a reidentification model usually performs much worse on unseen data. Partially this gap is caused by the relatively small scale of person re-identification datasets (compared to face recognition ones, for instance), but it is also related to training objectives. We propose to use the metric learning objective, namely AMSoftmax loss, and some additional training practices to build wellgeneralizing, yet, computationally efficient models. We use recently proposed Omni-Scale Network (OSNet) architecture combined with several training tricks and architecture adjustments to obtain state-of-the art results in cross-domain generalization problem on a large-scale MSMTI7 dataset in three setups: MSMTI7-all→DukeMTMC, MSMTI7-train→MarketI5OI and MSMTI7-all→MarketI5OI. Training code and the models are available online in the GitHub repository1.