摘要
小样本目标检测旨在通过少量的样本学习来训练目标检测模型,现有的小样本目标检测方法大多基于经典的目标检测算法。在二阶段的检测方法中,由于新类别样本数量少,产生了许多无关的边界框,导致候选区域的准确率较低。为了解决这个问题,提出了一种基于特征融合的小样本目标检测算法FF-FSOD。该方法采用特征融合的方法进行数据增强,对新类别样本进行补充,扩大样本的覆盖范围,同时引入FPN网络进行多尺度特征提取,再对RPN网络进行改进,引入支持集图像分支,计算支持集图像特征与查询集图像特征的深度互相关性,得到注意力特征图,进而获得更精确的候选框。所提模型的有效性在MS COCO和FSOD数据集上得到了验证,实验结果表明,该方法获得了更精准的候选框,进而提升了检测精度。
Few-shot object detection aims to train target detection model through a small amount of sample learning.At present,most of the existing few-shot object detection methods are based on classical target detection algorithms.In the two-stage detection method,due to the small number of new class samples,many irrelevant border boxes are generated,resulting in low accuracy of candidate regions.To solve this problem,this paper proposes a few-shot object detection algorithm FF-FSOD based on feature fusion.It uses the feature fusion method to enhance the data,supplements the new category samples,increases the coverage range of the sample,and introduces the FPN network to extract multi-scale feature.Then,the RPN network is improved,and the support set image branch is introduced.The depth correlation between the support set image feature and the query set image feature is calculated,and the attention feature map is obtained,and the more accurate candidate box is obtained.The effectiveness of the proposed model is verified on MS COCO and FSOD datasets.Experimental results show that the proposed method obtains more accurate candidate boxes and improves the detection accuracy.
作者
华杰
刘学亮
赵烨
HUA Jie;LIU Xueliang;ZHAO Ye(School of Computer Science and Information Engineering,Hefei University of Technology,Hefei 230601,China)
出处
《计算机科学》
CSCD
北大核心
2023年第2期209-213,共5页
Computer Science
基金
国家重点研发计划(2018AAA0102002)
国家自然科学基金(61976076,61632007)。
关键词
小样本学习
目标检测
深度学习
特征融合
特征金字塔
Few-shot learning
Object detection
Deep learning
Feature fusion
Feature pyramid
