R-FCN-3000, Segment Every Thing, YOLO9000

本文主要关注分类，检测和分割等任务训练数据的解耦，非常有趣。 R-FCN-3000 R-FCN-3000 obtains an mAP of 34.9% on the ImageNet detection dataset and outperforms YOLO 9000 by 18% while processing 30 images per second. 主要思路如上图，第一行将原来rfc

>>阅读原文<<