论文阅读
文章平均质量分 94
干饭的程序猿
这个作者很懒,什么都没留下…
展开
-
【论文阅读】Look-into-Object: Self-supervised Structure Modeling for Object Recognition
提出一个对象范围学习模块,用于根据同一类别中的实例之间共享的可视化模式定位对象。然后,我们设计了一个空间情境学习模块,通过预测物体在范围内的相对位置来建模物体的内部结构。这两个模块可以很容易地在训练期间插入任何骨干网,并在推理时分离。原创 2021-06-17 10:40:43 · 889 阅读 · 0 评论 -
【论文阅读】Look-into-Object: Self-supervised Structure Modeling for Object Recognition
【论文阅读】Look-into-Object: Self-supervised Structure Modeling for Object Recognition摘要具体实现分类模块对象范围学习模块空间背景学习模块摘要Most object recognition approaches predominantly focus on learning discriminative visual patterns while overlooking the holistic object structure原创 2021-05-20 16:16:23 · 447 阅读 · 0 评论 -
【论文阅读】Weakly Supervised Data Augmentation Network for Fine-Grained Visual Classification
【论文阅读】Weakly Supervised Data Augmentation Network for Fine-Grained Visual Classification摘要具体实现弱监督注意学习空间表示法双线性池化摘要Data augmentation is usually adopted to increase the amount of training data, prevent overfitting and improve the performance of deep models.原创 2021-05-14 15:58:33 · 696 阅读 · 0 评论 -
【论文阅读】BBN: Bilateral-Branch Network with Cumulative Learning for Long-Tailed Visual Recognition
【论文阅读】BBN: Bilateral-Branch Network with Cumulative Learning for Long-Tailed Visual Recognition摘要背景具体实现总体框架采样策略共享权重累计学习推理阶段消融实验代码实现(自己参照源码写的粗糙的网络)摘要Our work focuses on tackling the challenging but natural visual recognition task of long-tailed data distr原创 2021-04-13 10:23:38 · 1194 阅读 · 1 评论 -
【论文阅读】Learning Semantically Enhanced Feature for Fine-Grained Image Classification
[论文解读]Learning Semantically Enhanced Feature for Fine-Grained Image ClassificationAbstract本文旨在为细粒度图像分类(FGIC)提供一种计算量小但效果好的方法。与以往依赖复杂part定位模块的方法不同,我们的方法通过增强全局特征子特征的语义来学习细粒度特征。具体地说,我们首先通过通道排列将CNN的特征频道分成不同的组来实现子特征语义。同时,为了提高子特征的可区分性,通过加权组合正则化,引导分组在可区分性较强的obje原创 2021-04-05 09:09:11 · 1018 阅读 · 2 评论 -
【论文阅读】The Devil is in the Channels: Mutual-Channel Loss for Fine-Grained Image Classification
【论文阅读】The Devil is in the Channels: Mutual-Channel Loss for Fine-Grained Image Classification摘要具体实现判别组件差异性组件代码摘要Key for solving fine-grained image categorization is finding discriminate and local regions that correspond to subtle visual traits. Great str原创 2021-04-06 08:39:28 · 976 阅读 · 2 评论 -
【论文阅读】Cross-X Learning for Fine-Grained Visual Categorization
【论文阅读】Cross-X Learning for Fine-Grained Visual Categorization摘要具体实现OSME模块跨类别跨语义正则化(C3SC^{3} SC3S)跨层正则化(CLCLCL)摘要Recognizing objects from subcategories with very subtle differences remains a challenging task due to the large intra-class and small inter-cl原创 2021-04-06 22:01:32 · 1379 阅读 · 0 评论