Person re-identification (Re-ID) aims to retrieve specific individuals across non-overlapping camera views. In recent years, attention-based models contribute to many computer vision tasks due to their great ability for learning discriminative features. We propose the deep progressive attention (DPA) in a more natural manner for person Re-ID. Similar to human visual attention mechanism, the proposed DPA progressively selects the most discriminative parts of a specific individual and formulates feature representation for comparison purpose. Concretely, on the one hand, the proposed DPA uses a long-term reward to optimize the discriminative feature selection. On the other hand, a deep convolutional architecture is integrated into a recurrent model for feature representation learning. Extensive experiments on three person Re-ID benchmarks Market-1501, DukeMTMC-reID, and CUHK03-NP demonstrate the proposed DPA is on par with the state-of-the-art. Moreover, the experiments on partial person Re-ID datasets indicate the proposed DPA is competitive with the specially designed partial person Re-ID methods.

中文翻译：

---

对人员重新识别的深度渐进关注

人员重新识别 (Re-ID) 旨在通过非重叠的摄像机视图检索特定的个人。近年来，基于注意力的模型因其强大的学习判别特征的能力而对许多计算机视觉任务做出了贡献。我们以更自然的方式为行人 Re-ID 提出深度渐进关注 (DPA)。与人类视觉注意机制类似，所提出的 DPA 逐步选择特定个体最具辨别力的部分，并制定特征表示以进行比较。具体而言，一方面，所提出的 DPA 使用长期奖励来优化判别特征选择。另一方面，将深度卷积架构集成到用于特征表示学习的循环模型中。对三人 Re-ID 基准 Market-1501 的广泛实验，DukeMTMC-reID 和 CUHK03-NP 证明提议的 DPA 与最先进的技术相当。此外，在部分人 Re-ID 数据集上的实验表明，所提出的 DPA 与专门设计的部分人 Re-ID 方法具有竞争力。