| 深度细粒度图像识别研究综述 |
| |
| 引用本文： | 邓旭冉,闵少波,徐静远,李攀登,谢洪涛,张勇东.深度细粒度图像识别研究综述[J].南京气象学院学报,2019,11(6):625-637. |
| |
| 作者姓名： | 邓旭冉 闵少波 徐静远 李攀登 谢洪涛 张勇东 |
| |
| 作者单位： | 中国科学技术大学 信息科学技术学院, 合肥, 230026,中国科学技术大学 信息科学技术学院, 合肥, 230026,中国科学技术大学 信息科学技术学院, 合肥, 230026,中国科学技术大学 信息科学技术学院, 合肥, 230026,中国科学技术大学 信息科学技术学院, 合肥, 230026,中国科学技术大学 信息科学技术学院, 合肥, 230026 |
| |
| 基金项目： | 国家重点研发计划（2017YFC0820600） |
| |
| 摘 要： | 细粒度图像分类是计算机视觉中一项基础且重要的工作，其目的在于区分难以辨别的对象类别（例如不同子类的鸟类、花或动物）.不同于传统的图像分类任务可以雇佣大量普通人标注，细粒度数据集通常需要专家级知识进行标注.除了视觉分类中常见的姿态、光照和视角变化因素之外，细粒度数据集具有更大的类间相似性和类内差异性，因此要求模型能够捕捉到细微的类间差异信息和类内公有信息.除此之外，不同类别的样本存在不同程度的获取难度，因此细粒度数据集通常在数据分布中表现出长尾的特性.综上所述，细粒度数据分布具有小型、非均匀和不易察觉的类间差异等特点，对强大的深度学习算法也提出了巨大的挑战.本文首先介绍了细粒度图像分类任务的特点与挑战，随后以局部特征与全局特征两个主要视角整理了目前的主流工作，并讨论了它们的优缺点.最后在常用数据集上比较了相关工作的性能表现，并进行了总结与展望.
|
| 关 键 词： | 细粒度图像识别 深度学习 局部区域检测 双线性池化 |
| 收稿时间： | 2019/10/9 0:00:00 |
A survey of deep fine-grained visual categorization |
| |
| DENG Xuran,MIN Shaobo,XU Jingyuan,LI Pandeng,XIE Hongtao and ZHANG Yongdong.A survey of deep fine-grained visual categorization[J].Journal of Nanjing Institute of Meteorology,2019,11(6):625-637. |
| |
| Authors: | DENG Xuran MIN Shaobo XU Jingyuan LI Pandeng XIE Hongtao and ZHANG Yongdong |
| |
| Institution: | University of Science and Technology of China, School of Information Science and Technology, Hefei 230026,University of Science and Technology of China, School of Information Science and Technology, Hefei 230026,University of Science and Technology of China, School of Information Science and Technology, Hefei 230026,University of Science and Technology of China, School of Information Science and Technology, Hefei 230026,University of Science and Technology of China, School of Information Science and Technology, Hefei 230026 and University of Science and Technology of China, School of Information Science and Technology, Hefei 230026 |
| |
| Abstract: | Fine-grained image classification is a fundamental and important task in field of computer vision.The purpose of the task is to distinguish between object categories that have subtle inter-class differences (e.g.,birds,flowers,or animals of different sub-categories).Different from traditional image classification tasks that can employ a large number of common people for image annotations,fine-grained image classification usually requires expert-level knowledge.In addition to the common classification challenges of pose,lighting,and viewing changes,fine-grained datasets have larger inter-class similarity and intra-class variability.Therefore,it puts a high demand on the models to capture the subtle visual differences between classes and common intra-class characteristics.Furthermore,owing to the difficulty in obtaining samples of different categories,fine-grained datasets suffer from long-tail distribution problem.In summary,fine-grained data distribution has the characteristics of small,non-uniform,and indistinguishable inter-class differences,which also poses a huge challenge to the powerful deep learning algorithms.In this paper,we first introduce the formulation and challenges of fine-grained visual categorization tasks,and then illustrate two mainstream methods about local features and global features,as well as their advantages and disadvantages.Finally,we compare the performance of related works on common used datasets,and we make the required summarization and forecast. |
| |
| Keywords: | fine-grained visual categorization deep learning part region detection bilinear pooling |
|
| 点击此处可从《南京气象学院学报》浏览原始摘要信息 |
| 点击此处可从《南京气象学院学报》下载免费的PDF全文 |
|
