**作者：**Jian Ding, Nan Xue, Yang Long, Gui-Song Xia∗, Qikai Lu

发布时间：2019

发布期刊：CVPR

论文地址：

**论文全称：**Learning RoI Transformer for Oriented Object Detection in Aerial Images

代码：

https://github.com/dingjiansw101/AerialDetection

https://github.com/dingjiansw101/RoITransformer_DOTA

一、摘要

Object detection in aerial images is an active yet challenging task in computer vision because of the bird’s-eye view perspective, the highly complex backgrounds, and the variant appearances of objects. Especially when detecting densely packed objects in aerial images, methods relying on horizontal proposals for common object detection often introduce mismatches between the Region of Interests (RoIs) and objects. This leads to the common misalignment between the fifinal object classifification confifidence and localization accuracy. In this paper, we propose a RoI Transformer to address these problems. The core idea of RoI Transformer is to apply spatial transformations on RoIs and learn the transformation parameters under the supervision of oriented bounding box (OBB) annotations. RoI Transformer is with lightweight and can be easily embedded into detectors for oriented object detection. Simply apply the RoI Transformer to light-head RCNN has achieved state-of-the-art performances on two common and chal lenging aerial datasets, i.e*., DOTA and HRSC2016, with a neglectable reduction to detection speed. Our RoI Transformer exceeds the deformable Position Sensitive RoI pooling when oriented bounding-box annotations are available. Extensive experiments have also validated the flflexibility and effectiveness of our RoI Transformer.*

遥感图像目标检测的挑战：

• 遥感图像目标检测的困难：1. 鸟瞰图；2. 复杂的背景； 3. 多变的目标（主要指方向不确定——这一点最致命）
• 在对密集对象的检测时，使用依赖于生成水平候选框的检测算法，会导致ROI和目标之间的不匹配，即导致最后最终对象分类置信度和定位精度之间的不重合。

作者的解决方案：

• 提出使用ROI Transformer，即ROI形变——感兴趣区域的形变
• **核心思想：**在ROI上应用空间转变，并且学习在带有方向注释的bbox（OBB）上监督学习转变的参数

二、研究背景

2.1 遥感图像目标检测的困难

• 目标方向不确定——致命点
• 背景复杂
• 目标多变（形状、大小等）

2.2 HRoI 和 RRoI