欢迎访问光电子·激光编辑部网站！

您是第14631453位访问者！

熊炜,刘粤,许婷婷,孙鹏,赵迪,李利荣.损坏图像下基于风格归一化与全局注意力的行人重识别[J].光电子激光,2023,(8):833~841

损坏图像下基于风格归一化与全局注意力的行人重识别

Pedestrian re-identification based on style normalization and global attention in corrupted images

投稿时间：2022-07-28 修订日期：2022-11-21

DOI：

中文关键词: 行人重识别(ReID) 行人损坏图像平滑极大单元的风格归一化与恢复(SM-SNR) 全局注意力机制(GAM)

英文关键词:pedestrian re-identification (ReID) pedestrian corrupted image smooth maximum unit-style normalization and restitution (SM-SNR) module global attention mechanism (GAM)

基金项目:国家自然科学基金(61571182,61601177)、湖北省自然科学基金(2019CFB530)、湖北省科技厅重大专项 (2019ZYYD020)、襄阳湖北工业大学产业研究院科研项目(XYYJ2022C05)和国家留学基金(201808420418)资助项目

作者	单位
熊炜	湖北工业大学电气与电子工程学院，湖北武汉 430068 襄阳湖北工业大学产业研究院，湖北襄阳 441003 美国南卡罗来纳大学计算机科学与工程系，南卡罗来纳哥伦比亚 29201
刘粤	湖北工业大学电气与电子工程学院，湖北武汉 430068
许婷婷	湖北工业大学电气与电子工程学院，湖北武汉 430068
孙鹏	湖北工业大学电气与电子工程学院，湖北武汉 430068
赵迪	湖北工业大学电气与电子工程学院，湖北武汉 430068
李利荣	湖北工业大学电气与电子工程学院，湖北武汉 430068 襄阳湖北工业大学产业研究院，湖北襄阳 441003

摘要点击次数: 289

全文下载次数: 0

中文摘要:

针对当前网络难以应对各种损坏类型的行人图像与易丢失跨维信息的问题,提出了一种损坏图像下基于风格归一化与全局注意力的行人重识别(pedestrain re-identification,ReID) 方法。该方法通过平滑极大单元的风格归一化与恢复(smooth maximum unit-style normalization and restitution,SM-SNR) 模块中的实例规范化(instance normalization,IN) 过滤掉域中的风格变化,同时平滑极大单元(smooth maximum unit,SMU) 能使该模块更充分地从删除的信息中提取行人相关特征并将其恢复至网络中,缓解损坏图像带来的风格差异。此外,全局注意力机制(global attention mechanism,GAM) 通过关注通道与空间之间的相互作用,以捕获3个维度上的显著行人特征,减少跨维信息的丢失,最终使本模型在面对行人损坏图像时的识别能力得到有效提高,且保留了在干净数据集上的竞争力。实验结果表明,本算法在损坏测试集上的各项指标与目前主流算法对比具有显著的优越性。其中,本模型与2021年的CIL模型使用CUHK03数据集比较的结果为:在Corrupted Eval上,R-1、mAP和mINP分别提高了15.18%、15.75%与11.65%；在Clean Eval上,R-1与mINP仅降低了0.24%、0.75%,mAP提升了0.25%。

英文摘要:

Aiming at the problem that the current network is difficult to deal with various corrupted pedestrian images and easily loses cross-dimensional information,a pedestrian re-identification (ReID) method based on style normalization and global attention is proposed for corrupted images.The method filters out style changes in the domain by smooth maximum unit-style normalization and restitution (SM-SNR) module in the instance normalization (IN), and at the same time smooth maximum unit (SMU) enables the module to more fully extract pedestrian-related features from the deleted information and restore them to the network,so as to alleviate the style difference caused by corrupted images.In addition,the global attention mechanism (GAM) captures the salient pedestrian features in three dimensions by focusing on the interaction between the channel and the space,reducing the loss of cross-dimensional information.Finally,the recognition ability of the model in recognizing pedestrian corrupted images is effectively improved,and the competitiveness on clean datasets is retained.The experimental results show that the indicators of the algorithm on the corrupted test set has significant advantages compared with the current mainstream algorithms.Among these algorithms,the result of comparison with the 2021 CIL model using the CUHK03 dataset is that:On Corrupted Eval,R-1,mAP and mINP increase by 15.18%,15.75% and 11.65% respectively;on Clean Eval,R-1 and mINP only decrease by 0.24%,0.75%, and mAP increased by 0.25%.

查看全文下载PDF阅读器

关闭