Fine-Grained Visual Classification of Aircraft
Subhransu Maji TTI Chicago Esa Rahtu Juho Kannala University of Oulu, Finland {erahtu, Matthew Blaschko École Centrale Paris Andrea Vedaldi University of Oxford
Abstract
This paper introduces FGVC-Aircraft, a new dataset containing 10,000 images of aircraft spanning 100 aircraft models, organised in a three-level hierarchy. At the finer level, differences between models are often subtle but always visually measurable, making visual recognition challenging but possible. A benchmark is obtained by defining corresponding classification tasks and evaluation protocols, and baseline results are presented. The construction of this dataset was made possible by the work of aircraft enthusiasts, a strategy that can extend to the study of number of other object classes. Compared to the domains usually considered in fine-grained visual classification (FGVC), for example animals, aircraft are rigid and hence less deformable. They, however, present other interesting modes of variation, including purpose, size, designation, structure, historical style, and branding.
中文速览
细粒度视觉分类（Fine-Grained Visual Classification, FGVC）领域长期缺乏针对飞机的高质量基准数据集，FGVC-Aircraft 正是为填补这一空白而生。研究者从飞机观察爱好者（aircraft spotters）的在线摄影集中获取授权图片，构建了一个包含10,000张图像、覆盖100个飞机型号变体（variant）并按变体、家族（family）、制造商（manufacturer）三级层次组织的数据集，同时通过多样性最大化筛选和众包标注边界框来保证数据质量。以非线性SVM结合词袋（Bag-of-Visual-Words）特征为基线，型号变体识别准确率约为58.5%，制造商识别约为71.3%，说明数据集难度适中、挑战真实。这项工作不仅为FGVC引入了一个全新的刚性物体域，弥补了以往以动物为主的研究局限，其利用领域爱好者社区获取专业标注数据的方法论也为其他细粒度识别任务的数据集构建提供了可借鉴的范式。
原文 arXiv:1306.5151;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1306.5151v1