Abstract Multi-scale feature fusion has been proven effective in substantial person re-identification (ReID) works. However, the existing multi-scale feature fusion is based on features of different semantic levels. We propose a novel multi-scale and multi-branch feature representation for person ReID, namely Ms-Mb. It merges the features of the same semantic level and integrates attention modules to learn robust and representative feature representations. Through the heterogeneous losses supervision, the final feature representation of the image is more discriminative for person ReID. Sufficient ablation study has proven that the multi-scale feature fusion, the attention module and heterogeneous losses training strategy contribute to the performance boost of Ms-Mb. We have conducted experiments on four mainstream benchmarks including Market1501, DukeMTMC-reID, CUHK03 and MSMT17. Extensive experimental results show that our approach Ms-Mb achieves state-of-the-art performances on Market1501 (Rank-1 = 95.8 % , mAP = 89.9 % ), DukeMTMC-reID (Rank-1 = 90.8 % , mAP = 82.2 % ) and MSMT17 (Rank-1 = 81.9 % , mAP = 59.3 % ) without using additional external data or re-ranking. Our approach yields competitive results compared to the state-of-the-art method on CUHK03 (Rank-1 = 75.4 % , mAP = 72.9 % ) and surpasses the other methods by a large margin.

中文翻译：

---

用于人员重新识别的多尺度和多分支特征表示

摘要 多尺度特征融合已被证明在大量人员重新识别（ReID）工作中是有效的。然而，现有的多尺度特征融合是基于不同语义层次的特征。我们为行人 ReID 提出了一种新颖的多尺度和多分支特征表示，即 Ms-Mb。它合并了相同语义级别的特征，并集成了注意力模块来学习鲁棒的和有代表性的特征表示。通过异构损失监督，图像的最终特征表示对行人 ReID 更具辨别力。充分的消融研究已经证明，多尺度特征融合、注意力模块和异构损失训练策略有助于提高 Ms-Mb 的性能。我们在四个主流基准上进行了实验，包括 Market1501、DukeMTMC-reID、CUHK03 和 MSMT17。大量实验结果表明，我们的方法 Ms-Mb 在 Market1501 (Rank-1 = 95.8 % , mAP = 89.9 % )、DukeMTMC-reID (Rank-1 = 90.8 % , mAP = 82.2 %) 上取得了最先进的性能) 和 MSMT17 (Rank-1 = 81.9 % , mAP = 59.3 % ) 而不使用额外的外部数据或重新排名。与 CUHK03 上的最新方法（Rank-1 = 75.4 % , mAP = 72.9 % ）相比，我们的方法产生了有竞争力的结果，并且大大超过了其他方法。