Knowledge Commons of Institute of Automation,CAS
Multi-Resolution and Semantic-Aware Bidirectional Adapter for Multi-Scale Object Detection | |
Li, Zekun1; Pan, Jin1; He, Peidong2,3; Zhang, Ziqi4; Zhao, Chunlu1; Li, Bing4 | |
发表期刊 | APPLIED SCIENCES-BASEL |
2023-12-01 | |
卷号 | 13期号:23页码:19 |
通讯作者 | Zhao, Chunlu(chunluzhao@cert.org.cn) |
摘要 | Scale variation presents a significant challenge in object detection. To address this, multi-level feature fusion techniques have been proposed, exemplified by methods such as the feature pyramid network (FPN) and its extensions. Nonetheless, the input features provided to these methods and the interaction among features across different levels are limited and inflexible. In order to fully leverage the features of multi-scale objects and amplify feature interaction and representation, we introduce a novel and efficient framework known as a multi-resolution and semantic-aware bidirectional adapter (MSBA). Specifically, MSBA comprises three successive components: multi-resolution cascaded fusion (MCF), a semantic-aware refinement transformer (SRT), and bidirectional fine-grained interaction (BFI). MCF adaptively extracts multi-level features to enable cascaded fusion. Subsequently, SRT enriches the long-range semantic information within high-level features. Following this, BFI facilitates ample fine-grained interaction via bidirectional guidance. Benefiting from the coarse-to-fine process, we can acquire robust multi-scale representations for a variety of objects. Each component can be individually integrated into different backbone architectures. Experimental results substantiate the superiority of our approach and validate the efficacy of each proposed module. |
关键词 | object detection scale variation transformer multi-level fusion |
DOI | 10.3390/app132312639 |
收录类别 | SCI |
语种 | 英语 |
资助项目 | National Natural Science Foundation of China |
项目资助者 | National Natural Science Foundation of China |
WOS研究方向 | Chemistry ; Engineering ; Materials Science ; Physics |
WOS类目 | Chemistry, Multidisciplinary ; Engineering, Multidisciplinary ; Materials Science, Multidisciplinary ; Physics, Applied |
WOS记录号 | WOS:001118020600001 |
出版者 | MDPI |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/55049 |
专题 | 多模态人工智能系统全国重点实验室 |
通讯作者 | Zhao, Chunlu |
作者单位 | 1.Coordinat Ctr China CNCERT CC, Natl Comp Network Emergency Response Tech Team, Beijing 100029, Peoples R China 2.Chinese Acad Sci, Aerosp Informat Res Inst, 9 Dengzhuang South Rd, Beijing 100094, Peoples R China 3.Chinese Acad Sci, Dept Key Lab Computat Opt Imaging Technol, 9 Dengzhuang South Rd, Beijing 100094, Peoples R China 4.Chinese Acad Sci, Inst Automat, State Key Lab Multimodal Artificial Intelligence S, Beijing 100094, Peoples R China |
推荐引用方式 GB/T 7714 | Li, Zekun,Pan, Jin,He, Peidong,et al. Multi-Resolution and Semantic-Aware Bidirectional Adapter for Multi-Scale Object Detection[J]. APPLIED SCIENCES-BASEL,2023,13(23):19. |
APA | Li, Zekun,Pan, Jin,He, Peidong,Zhang, Ziqi,Zhao, Chunlu,&Li, Bing.(2023).Multi-Resolution and Semantic-Aware Bidirectional Adapter for Multi-Scale Object Detection.APPLIED SCIENCES-BASEL,13(23),19. |
MLA | Li, Zekun,et al."Multi-Resolution and Semantic-Aware Bidirectional Adapter for Multi-Scale Object Detection".APPLIED SCIENCES-BASEL 13.23(2023):19. |
条目包含的文件 | 条目无相关文件。 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论