Abstract
Dense small objects in UAV imagery are often missed due to long-range viewpoints, occlusion, and clutter[cite: 5]. This paper presents a detector-agnostic post-processing framework that converts overlap-induced redundancy into group evidence[cite: 6]. Overlapping tiling first recovers low-confidence candidates[cite: 7]. A Spatial Gate (DBSCAN on box centroids) and a Semantic Gate (DBSCAN on ResNet-18 embeddings) then validates group evidence[cite: 7]. Validated groups receive controlled confidence reweighting before class-aware NMS fusion[cite: 8]. Experiments on VisDrone show a recall increase from 0.685 to 0.778 (+0.093) and a precision adjustment from 0.801 to 0.595, yielding F1=0.669[cite: 9]. Post-processing latency averages 0.095 s per image[cite: 10]. These results indicate recall-first, precision-trade-off behavior that benefits recall-sensitive applications such as far-field counting and monitoring[cite: 10]. Ablation confirms that tiling exposes missed objects, spatial clustering stabilizes geometry, semantic clustering enforces appearance coherence, and reweighting provides calibrated integration with the baseline[cite: 11]. The framework requires no retraining and integrates with modern detectors[cite: 12]. Future work will reduce semantic gating cost and extend the approach with temporal cues[cite: 13].
Abstract (translated)
无人机(UAV)图像中的密集小物体由于长距离视角、遮挡和杂乱背景等原因常常被遗漏[cite: 5]。本文提出了一种与检测器无关的后处理框架,该框架将由重叠引起的冗余转换为群体证据[cite: 6]。首先通过重叠分块恢复低置信度候选物体[cite: 7]。然后使用空间门(基于目标框中心点进行DBSCAN聚类)和语义门(基于ResNet-18特征嵌入进行DBSCAN聚类)来验证群体证据的有效性[cite: 7]。经过验证的群体在置信度重新加权后,再通过类别感知的非极大值抑制(NMS)融合处理[cite: 8]。实验结果表明,在VisDrone数据集上的召回率从0.685提升至0.778 (+0.093),同时精度调整为0.595,得到了F1分数为0.669的表现[cite: 9]。后处理延迟平均为每张图像0.095秒[cite: 10]。这些结果显示了以召回率为先、牺牲部分精确度的特性,这在远距离计数和监控等需要高召回率的应用中尤为有益[cite: 10]。消融实验确认了分块可以发现被遗漏的对象,空间聚类可稳定几何关系,语义聚类则能确保外观一致性,并且重新加权提供了与基线模型的校准融合效果[cite: 11]。该框架无需重新训练即可与现代检测器集成使用[cite: 12]。未来的工作将致力于降低语义门的成本,并通过引入时间线索来扩展此方法的应用范围[cite: 13]。
URL
https://arxiv.org/abs/2509.10779