论文标题
对齐对象检测的深度功能
Align Deep Features for Oriented Object Detection
论文作者
论文摘要
在过去的十年中，在检测空中图像中通常以大规模变化和任意方向分配的对象方面取得了重大进展。但是，大多数现有方法都依赖于具有不同尺度，角度和纵横比的启发式定义的锚，并且通常在锚固箱和与轴平均的卷积特征之间严重损坏，这会导致分类分数和定位得分之间的常见不一致。为了解决此问题，我们提出了一个由两个模块组成的单拍对网络（S $^2 $ a-net）：一个功能对齐模块（FAM）和一个方向的检测模块（ODM）。 FAM可以通过锚固网络产生高质量的锚点，并根据锚固框与新型的对齐卷积自适应地对齐卷积特征。 ODM首先采用主动旋转过滤器来编码方向信息，然后产生方向敏感和定向不变的特征，以减轻分类得分和本地化精度之间的不一致。此外，我们进一步探讨了检测大尺寸图像中对象的方法，从而可以在速度和准确性之间进行更好的权衡。广泛的实验表明，我们的方法可以在两个常用的空中对象数据集（即DOTA和HRSC2016）上实现最先进的性能，同时保持高效率。该代码可在https://github.com/csuhan/s2anet上找到。
The past decade has witnessed significant progress on detecting objects in aerial images that are often distributed with large scale variations and arbitrary orientations. However most of existing methods rely on heuristically defined anchors with different scales, angles and aspect ratios and usually suffer from severe misalignment between anchor boxes and axis-aligned convolutional features, which leads to the common inconsistency between the classification score and localization accuracy. To address this issue, we propose a Single-shot Alignment Network (S$^2$A-Net) consisting of two modules: a Feature Alignment Module (FAM) and an Oriented Detection Module (ODM). The FAM can generate high-quality anchors with an Anchor Refinement Network and adaptively align the convolutional features according to the anchor boxes with a novel Alignment Convolution. The ODM first adopts active rotating filters to encode the orientation information and then produces orientation-sensitive and orientation-invariant features to alleviate the inconsistency between classification score and localization accuracy. Besides, we further explore the approach to detect objects in large-size images, which leads to a better trade-off between speed and accuracy. Extensive experiments demonstrate that our method can achieve state-of-the-art performance on two commonly used aerial objects datasets (i.e., DOTA and HRSC2016) while keeping high efficiency. The code is available at https://github.com/csuhan/s2anet.
