Small object detection requires the detection head to scan a large number of positions on image feature maps, which is extremely hard for computation- and energy-efficient lightweight generic detectors. To accurately detect small objects with limited computation, we propose a two-stage lightweight detection framework with extremely low computation complexity, termed as TinyDet. It enables high-resolution feature maps for dense anchoring to better cover small objects, proposes a sparsely-connected convolution for computation reduction, enhances the early stage features in the backbone, and addresses the feature misalignment problem for accurate small object detection. On the COCO benchmark, our TinyDet-M achieves 30.3 AP and 13.5 AP^s with only 991 MFLOPs, which is the first detector that has an AP over 30 with less than 1 GFLOPs; besides, TinyDet-S and TinyDet-L achieve promising performance under different computation limitation.

翻译：小目标检测需要检测头在图像特征图上扫描大量位置，这对于计算和能效受限的轻量级通用检测器而言极为困难。为在有限计算资源下精确检测小目标，我们提出了一种计算复杂度极低的两阶段轻量级检测框架，称为TinyDet。该框架通过高分辨率特征图实现密集锚点以更好地覆盖小目标，提出稀疏连接卷积以降低计算量，增强骨干网络早期阶段特征，并解决特征未对齐问题以实现精确的小目标检测。在COCO基准测试中，我们的TinyDet-M在仅991 MFLOPs计算量下达到30.3 AP和13.5 AP^s，这是首个在小于1 GFLOPs计算量下AP超过30的检测器；此外，TinyDet-S和TinyDet-L在不同计算限制下均取得了有竞争力的性能。