SCRDet: Towards More Robust Detection for Small, Cluttered and Rotated Objects

TL;DR

SCRDet is a novel multi-category rotation detector designed to improve detection of small, cluttered, and rotated objects, especially in aerial images, by fusing multi-layer features and employing attention mechanisms for enhanced sensitivity and accuracy.

Contribution

The paper introduces SCRDet, a new detector combining multi-layer feature fusion, attention mechanisms, and an IoU-based loss for better detection of challenging objects in various datasets.

Findings

Achieves state-of-the-art results on remote sensing datasets DOTA and NWPU VHR-10.

Outperforms existing methods on natural image datasets COCO and VOC2007.

Demonstrates robustness in detecting small, cluttered, and rotated objects.

Abstract

Object detection has been a building block in computer vision. Though considerable progress has been made, there still exist challenges for objects with small size, arbitrary direction, and dense distribution. Apart from natural images, such issues are especially pronounced for aerial images of great importance. This paper presents a novel multi-category rotation detector for small, cluttered and rotated objects, namely SCRDet. Specifically, a sampling fusion network is devised which fuses multi-layer feature with effective anchor sampling, to improve the sensitivity to small objects. Meanwhile, the supervised pixel attention network and the channel attention network are jointly explored for small and cluttered object detection by suppressing the noise and highlighting the objects feature. For more accurate rotation estimation, the IoU constant factor is added to the smooth L1 loss to address the boundary problem for the rotating bounding box. Extensive experiments on two remote sensing public datasets DOTA, NWPU VHR-10 as well as natural image datasets COCO, VOC2007 and scene text data ICDAR2015 show the state-of-the-art performance of our detector. The code and models will be available at https://github.com/DetectionTeamUCAS.