arXiv 2407.010162025-09-25cs.CV

SOOD++：利用未标记数据提升定向目标检测

SOOD++: Leveraging Unlabeled Data to Boost Oriented Object Detection

School of Artificial Intelligence and Automation, Huazhong University of Science and Technology（人工智能与自动化学院，华中科技大学）；School of Electronic Information and Communications, Huazhong University of Science and Technology（电子信息与通信学院，华中科技大学）；Baidu Inc., China（百度公司，中国）；School of Software Engineering, Huazhong University of Science and Technology（软件工程学院，华中科技大学）

Dingkang Liang, Wei Hua, Chunsheng Shi, Zhikang Zou, Xiaoqing Ye, Xiang Bai

备注：: Accepted by IEEE TPAMI. The project page is at https://dk-liang.github.io/SOODv2/

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对航空图像中定向目标检测的标注成本高问题，提出半监督方法SOOD++，通过密集伪标签、几何感知加权和全局一致性，在多个基准上显著超越现有方法。

摘要：

半监督目标检测（SSOD）利用未标记数据来提升目标检测器性能，近年来已成为热门研究课题。然而，现有的SSOD方法主要关注水平目标，而航空图像中常见的定向目标尚未得到充分探索。同时，定向目标的标注成本显著高于水平目标。因此，本文提出了一种简单而有效的半监督定向目标检测方法，称为SOOD++。具体而言，我们观察到航空图像中的目标通常具有任意方向、小尺度和密集分布的特点，这启发了以下核心设计：采用简单实例感知密集采样（SIDS）策略生成全面的密集伪标签；几何感知自适应加权（GAW）损失通过利用航空目标的复杂几何信息，动态调整每个伪标签与对应预测之间配对的重要性；我们将航空图像视为全局布局，并通过提出的噪声驱动全局一致性（NGC）显式构建伪标签集合与预测集合之间的多对多关系。在各种定向目标数据集和不同标注设置下进行的大量实验证明了我们方法的有效性。例如，在DOTA-V2.0/DOTA-V1.5基准上，在10%、20%和30%标注数据设置下，采用单尺度训练和测试，所提方法分别以较大优势（+2.90/2.14、+2.16/2.18和+2.66/2.32）mAP超越了先前的最先进（SOTA）方法。更重要的是，在利用完整DOTA-V1.5训练验证集训练的强监督基线（70.66 mAP）基础上，该方法仍提升了+1.82 mAP，达到72.48 mAP，推动了新的最先进水平。项目页面位于https://dk-liang.github.io/SOODv2/。

英文摘要：

Semi-supervised object detection (SSOD), leveraging unlabeled data to boost object detectors, has become a hot topic recently. However, existing SSOD approaches mainly focus on horizontal objects, leaving oriented objects common in aerial images unexplored. At the same time, the annotation cost of oriented objects is significantly higher than that of their horizontal counterparts. Therefore, in this paper, we propose a simple yet effective Semi-supervised Oriented Object Detection method termed SOOD++. Specifically, we observe that objects from aerial images usually have arbitrary orientations, small scales, and dense distribution, which inspires the following core designs: a Simple Instance-aware Dense Sampling (SIDS) strategy is used to generate comprehensive dense pseudo-labels; the Geometry-aware Adaptive Weighting (GAW) loss dynamically modulates the importance of each pair between pseudo-label and corresponding prediction by leveraging the intricate geometric information of aerial objects; we treat aerial images as global layouts and explicitly build the many-to-many relationship between the sets of pseudo-labels and predictions via the proposed Noise-driven Global Consistency (NGC). Extensive experiments conducted on various oriented object datasets under various labeled settings demonstrate the effectiveness of our method. For example, on the DOTA-V2.0/DOTA-V1.5 benchmark, the proposed method outperforms previous state-of-the-art (SOTA) by a large margin (+2.90/2.14, +2.16/2.18, and +2.66/2.32) mAP under 10%, 20%, and 30% labeled data settings, respectively, with single-scale training and testing. More importantly, it still improves upon a strong supervised baseline with 70.66 mAP, trained using the full DOTA-V1.5 train-val set, by +1.82 mAP, resulting in a 72.48 mAP, pushing the new state-of-the-art. The project page is at https://dk-liang.github.io/SOODv2/

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。