Zou Wenbin, Zhang Zhengyu, Peng Yingqing, Xiang Canqun, Tian Shishun, Zhang Lu
IEEE Trans Image Process. 2021;30:4084-4098. doi: 10.1109/TIP.2021.3069547. Epub 2021 Apr 8.
Current state-of-the-art two-stage detectors heavily rely on region proposals to guide the accurate detection for objects. In previous region proposal approaches, the interaction between different functional modules is correlated weakly, which limits or decreases the performance of region proposal approaches. In this paper, we propose a novel two-stage strong correlation learning framework, abbreviated as SC-RPN, which aims to set up stronger relationship among different modules in the region proposal task. Firstly, we propose a Light-weight IoU-Mask branch to predict intersection-over-union (IoU) mask and refine region classification scores as well, it is used to prevent high-quality region proposals from being filtered. Furthermore, a sampling strategy named Size-Aware Dynamic Sampling (SADS) is proposed to ensure sampling consistency between different stages. In addition, point-based representation is exploited to generate region proposals with stronger fitting ability. Without bells and whistles, SC-RPN achieves AR 14.5% higher than that of Region Proposal Network (RPN), surpassing all the existing region proposal approaches. We also integrate SC-RPN into Fast R-CNN and Faster R-CNN to test its effectiveness on object detection task, the experimental results achieve a gain of 3.2% and 3.8% in terms of mAP compared to the original ones.
当前最先进的两阶段检测器严重依赖区域提议来指导目标的精确检测。在以往的区域提议方法中，不同功能模块之间的交互关联较弱，这限制或降低了区域提议方法的性能。在本文中，我们提出了一种新颖的两阶段强关联学习框架，简称为SC-RPN，旨在在区域提议任务中建立不同模块之间更强的关系。首先，我们提出了一个轻量级IoU掩码分支来预测交并比（IoU）掩码并细化区域分类分数，它用于防止高质量区域提议被过滤。此外，还提出了一种名为大小感知动态采样（SADS）的采样策略，以确保不同阶段之间的采样一致性。此外，利用基于点的表示来生成具有更强拟合能力的区域提议。在没有花里胡哨功能的情况下，SC-RPN的平均召回率（AR）比区域提议网络（RPN）高14.5%，超过了所有现有的区域提议方法。我们还将SC-RPN集成到快速R-CNN和更快R-CNN中，以测试其在目标检测任务上的有效性，实验结果在平均精度均值（mAP）方面比原来的方法分别提高了3.2%和3.8%。
