目录： 2016[CVPR]-Embedding Label Structures for Fine-Grained Feature Representation 2016[CVPR]-Learning Deep Representations of Fine-Grained Visual Descriptions 2016[CVPR]-Part-Stacked CNN for Fine-Grained Visual Categorization 2017[CVPR]-Low-rank Bilinear Pooling for Fine-Grained Classification 2017[CVPR]-Mining Discriminative Triplets of Patches for Fine-Grained Classification 2017[ICCV]-Efficient Fine-grained Classification and Part Localization Using One Compact Network 2017[ICCV]-Fine-grained recognition in the wild A multi-task domain adaptation approach 2017[TMM]-Diversified visual attention networks for fine-grained object classification
摘要：细粒度的物体分类需要精确的差异区域的定位。本文提出了一种patch-based的框架来解决这个问题。文章使用patchs的几何约束的triplets loss来优化定位的准确度。并且使用一个自动的自我其实几何约束triplets 用于分类。（真不知道中文咋翻）
。。。。这个系列太无聊了，我弃坑了，不更新了不更新了。。。这篇稿子放了快一个月了还没写完
