arrow
返回

Aspect-Ratio Aware Dynamic Label Assignment for Oriented Object Detection

delete2024-01-01
delete0
PRE
AI
C
Congcong Wei
魏丹 封面图
魏丹 (Dan Wei) *
DOI:10.1109/LGRS.2024.3432810delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Due to the presence of objects with large aspect ratios and variations in their angles, the use of oriented object detectors for remote sensing images faces significant challenges. As a result, detection strategies for oriented object detectors have been extensively studied. The label assignment strategy and angle loss calculation often need to consider the objects' shape characteristics fully within these detectors. A small deviation in the angle prediction leads to a dramatic drop in intersection over union (IoU) between the prediction box and the ground-truth (GT) box. This can lead to problems with large aspect ratio objects in the presence of insufficient sampling and loss misalignment. To address these issues, this letter proposes a new and flexible aspect-ratio aware (ARA) dynamic label assignment strategy and aligned angle loss (AAL) strategy for oriented object detection. Precisely, we utilize information such as the prediction classification score, the IoU between the prediction boxes and the GT boxes, and the aspect ratio of the GT boxes to evaluate the quality of the candidate samples to dynamically select positive and negative samples to improve the quality of the positive samples. In addition, to further enhance the detection capability for objects with large aspect ratios, we decouple the localization loss and incorporate aspect ratio information into the loss calculation, thereby aligning angle loss for objects with varying aspect ratios. Extensive experiments on popular remote sensing datasets, such as DOTA and HRSC2016, validate the effectiveness and superiority of our proposed method.
Keyword:
Angle loss
dynamic label assignment
object detection
Angle loss
dynamic label assignment
object detection

期刊

IEEE Geoscience and Remote Sensing Magazine 封面图
IEEE Geoscience and Remote Sensing Magazine
IF:
16.4
论文数:
1.0W
被引数:
5.1K

机构

S
Shanghai University of Engineering Science
学者数:
7.9K
论文数: 4.9K
被引数: 6.0K
引用论文

引用论文

err分享
err收藏
Inhibition of autophagy augments chemotherapy in human salivary adenoid cystic carcinoma
err2013-10-09
err0
PREAI
errLicheng Jiang; Shengyun Huang; Dongsheng Zhang; Bin Zhang; Keyi Li; Wengang Li; Shizhou Zhang; Weidong Zhang; Peihui Zheng
err分享
err收藏
Frontiers in spin-polarized tunneling
err2010-04-01
err0
PREAI
errJagadeesh S. Moodera; Guo-Xing Miao; Tiffany S. Santos
err分享
err收藏
Oriented R-CNN and Beyond面向r-cnn及以后
err2024-01-29
err9
PREAI
errXie, Xingxing; Cheng, Gong; Wang, Jiabao; Li, Ke; Yao, Xiwen; Han, Junwei
err分享
err收藏
学者 查看更多内容