Point RCNN: An Angle-Free Framework for Rotated Object Detection

TL;DR

Point RCNN introduces an angle-free rotated object detection framework that improves accuracy and stability in aerial images by using representative points and re-sampling rare categories, achieving state-of-the-art results.

Contribution

The paper proposes a novel angle-free detection framework, Point RCNN, with a coarse-to-fine RRoI generation and corner refinement, addressing boundary issues and data imbalance.

Findings

Achieves state-of-the-art results on DOTA datasets.

Effectively handles boundary problems in angle regression.

Improves detection stability with re-sampling of rare categories.

Abstract

Rotated object detection in aerial images is still challenging due to arbitrary orientations, large scale and aspect ratio variations, and extreme density of objects. Existing state-of-the-art rotated object detection methods mainly rely on angle-based detectors. However, angle regression can easily suffer from the long-standing boundary problem. To tackle this problem, we propose a purely angle-free framework for rotated object detection, called Point RCNN, which mainly consists of PointRPN and PointReg. In particular, PointRPN generates accurate rotated RoIs (RRoIs) by converting the learned representative points with a coarse-to-fine manner, which is motivated by RepPoints. Based on the learned RRoIs, PointReg performs corner points refinement for more accurate detection. In addition, aerial images are often severely unbalanced in categories, and existing methods almost ignore this issue. In this paper, we also experimentally verify that re-sampling the images of the rare categories will stabilize training and further improve the detection performance. Experiments demonstrate that our Point RCNN achieves the new state-of-the-art detection performance on commonly used aerial datasets, including DOTA-v1.0, DOTA-v1.5, and HRSC2016.