混合采样下多级特征聚合的视频目标检测算法
秦思怡,盖绍彦,达飞鹏

Video object detection algorithm based on multi-level feature aggregation under mixed sampler
Siyi QIN,Shaoyan GAI,Feipeng DA
表 4 多级特征聚合注意力模块在ImageNet VID验证集上不同速度目标的精度
Tab.4 Accuracy of detecting objects with different speeds of MFAA module on ImageNet VID verification set
%
速度 AP50
YOLOX-S YOLOX-S+SA YOLOX-S+MFAA
慢速 80.1 81.8 81.9
中速 71.4 75.4 75.6
快速 55.3 58.3 59.3
平均精度 69.5 76.9 77.5