1. INTRODUCTION
Person detection is an important topic in the computer vision area. It has wide applications including surveillance [1][2], driving assistance [3] and maritime quick rescue [4], etc. The research of detectors [5][6][7][8][9] has achieved significant progress with the rapid development of data-driven deep convolutional neural networks (CNNs). However, the detectors perform poorly when detecting tiny objects with few pixels (e.g., less than 20 × 20 pixels), such as traffic signs [10], persons in aerial images [4], etc.