期刊
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
卷 23, 期 12, 页码 25458-25468出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TITS.2021.3127632
关键词
Semantics; Task analysis; Visualization; Training; Feature extraction; Autonomous vehicles; Real-time systems; Autonomous driving; few-shot image classification; zero-shot image classification; attention mechanism; visual-semantic alignment
资金
- Sichuan Science and Technology Program, China [2019ZDZX0008, 2019YFG0003, 2019YFG0533]
- National Natural Science Foundation of China [61976049, 62072080, U20B2063]
- Fundamental Research Funds for the Central Universities [ZYGX2019Z015]
Autonomous driving relies on accurate visual recognition of surrounding objects, and few-shot image classification is employed to recognize rarely seen objects. This paper introduces a Semantic-Aligned Attention (SAA) mechanism to refine feature embedding in existing embedding and metric-learning approaches, showing competitive improvements in both few-shot and zero-shot classification tasks on benchmark datasets.
Autonomous driving relies on trusty visual recognition of surrounding objects. Few-shot image classification is used in autonomous driving to help recognize objects that are rarely seen. Successful embedding and metric-learning approaches to this task normally learn a feature comparison framework between an unseen image and the labeled images. However, these approaches usually have problems with ambiguous feature embedding because they tend to ignore important local visual and semantic information when extracting intra-class common features from the images. In this paper, we introduce a Semantic-Aligned Attention (SAA) mechanism to refine feature embedding and it can be applied to most of the existing embedding and metric-learning approaches. The mechanism highlights pivotal local visual information with attention mechanism and aligns the attentive map with semantic information to refine the extracted features. Incorporating the proposed mechanism into the prototypical network, evaluation results reveal competitive improvements in both few-shot and zero-shot classification tasks on various benchmark datasets.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据