论文标题
在无锚对象检测中降低标签噪声
Reducing Label Noise in Anchor-Free Object Detection
论文作者
论文摘要
当前的无锚对象检测器将空间落在地面框的预定义中心区域内的所有特征标记为正。这种方法会在训练过程中引起标签噪声，因为这些正面标记的特征中的某些可能在后台或封锁对象上，或者它们根本不是歧视性特征。在本文中，我们提出了一种新的标签策略，旨在减少无锚探测器中的标签噪声。我们将来自单个特征的预测汇总到单个预测中。这使模型可以减少训练期间非歧视性特征的贡献。我们开发了一种新的单阶段，无锚的对象检测器PPDET，以在训练过程中采用此标签策略，并在推理过程中采用类似的预测池方法。在可可数据集上，PPDET在无锚自上而下的探测器中实现了最佳性能，并使用其他最先进的方法进行了PAR。在小对象检测中，它还优于所有主要的单阶段和两阶段方法（$ {ap} _ {s} $ 31.4 $）。代码可从https://github.com/nerminsamet/ppdet获得
Current anchor-free object detectors label all the features that spatially fall inside a predefined central region of a ground-truth box as positive. This approach causes label noise during training, since some of these positively labeled features may be on the background or an occluder object, or they are simply not discriminative features. In this paper, we propose a new labeling strategy aimed to reduce the label noise in anchor-free detectors. We sum-pool predictions stemming from individual features into a single prediction. This allows the model to reduce the contributions of non-discriminatory features during training. We develop a new one-stage, anchor-free object detector, PPDet, to employ this labeling strategy during training and a similar prediction pooling method during inference. On the COCO dataset, PPDet achieves the best performance among anchor-free top-down detectors and performs on-par with the other state-of-the-art methods. It also outperforms all major one-stage and two-stage methods in small object detection (${AP}_{S}$ $31.4$). Code is available at https://github.com/nerminsamet/ppdet
