论文标题
SFPN：用于对象检测的合成FPN
SFPN: Synthetic FPN for Object Detection
论文作者
论文摘要
FPN（特征金字塔网络）已成为大多数SOTA一个阶段对象检测器的基本组成部分。以前的许多研究反复证明，FPN可以更好地多尺度特征图，如果对象的大小不同，可以更精确地描述对象。但是，对于大多数骨架，例如VGG，Resnet或densenet，由于合并操作或相对于步幅为2的卷积，每层的特征图都缩小到其四分之一。下缩放by-2的间隙很大，并使其FPN的fpn融合不平稳。本文提出了一个新的SFPN（合成融合金字塔网络），该杂质在原始FPN的层之间创建了各种合成层，以增强轻质CNN后台的准确性，以更准确地提取对象的视觉特征。最后，实验证明SFPN体系结构的表现优于大型骨干VGG16，RESNET50或基于AP得分等轻量重量骨架，例如MobilenetV2。
FPN (Feature Pyramid Network) has become a basic component of most SoTA one stage object detectors. Many previous studies have repeatedly proved that FPN can caputre better multi-scale feature maps to more precisely describe objects if they are with different sizes. However, for most backbones such VGG, ResNet, or DenseNet, the feature maps at each layer are downsized to their quarters due to the pooling operation or convolutions with stride 2. The gap of down-scaling-by-2 is large and makes its FPN not fuse the features smoothly. This paper proposes a new SFPN (Synthetic Fusion Pyramid Network) arichtecture which creates various synthetic layers between layers of the original FPN to enhance the accuracy of light-weight CNN backones to extract objects' visual features more accurately. Finally, experiments prove the SFPN architecture outperforms either the large backbone VGG16, ResNet50 or light-weight backbones such as MobilenetV2 based on AP score.
