qq_42820467的博客

参考论文：Visible Thermal Person Re-Identification via Dual-Constrained Top-RankingBi-Directional Center-Constrained Top-Ranking for Visible Thermal Person Re-IdentificationBaseline 采用Resnet50，由两个主要部分组成：用于特征提取的双路径网络（一条用于可见光图像，另一条用于热图像）和用于特征学习的双向双约束顶级损失。这里先

attentionseq2seq-models-with-attentionhttps://jalammar.github.io/visualizing-neural-machine-translation-mechanics-of-seq2seq-models-with-attention/Transformer在视觉领域的应用https://www.cnblogs.com/ziyuzhu-edward/p/14132194.htmltransformer模型结构http://jalammar