期刊
INTERNATIONAL JOURNAL OF COMPUTER VISION
卷 131, 期 3, 页码 772-795出版社
SPRINGER
DOI: 10.1007/s11263-022-01731-4
关键词
Machine learning; Few-shot learning; Meta learning; Point cloud classification
Few-shot learning (FSL) has gained rapid growth in the 2D image domain due to less labeled training data requirement and better generalization, but its application in 3D point cloud data is relatively unexplored. 3D FSL is more challenging with irregular structures, subtle inter-class differences, and high intra-class variances. This work performs systematic investigations of applying recent 2D FSL works to 3D point cloud backbone networks and proposes a new network, Point-cloud Correlation Interaction (PCIA), with novel plug-and-play components to improve feature distinction. Experimental results demonstrate our method achieves state-of-the-art performance for 3D FSL.
In recent years, research on few-shot learning (FSL) has been fast-growing in the 2D image domain due to the less requirement for labeled training data and greater generalization for novel classes. However, its application in 3D point cloud data is relatively under-explored. Not only need to distinguish unseen classes as in the 2D domain, 3D FSL is more challenging in terms of irregular structures, subtle inter-class differences, and high intra-class variances when trained on a low number of data. Moreover, different architectures and learning algorithms make it difficult to study the effectiveness of existing 2D FSL algorithms when migrating to the 3D domain. In this work, for the first time, we perform systematic and extensive investigations of directly applying recent 2D FSL works to 3D point cloud related backbone networks and thus suggest a strong learning baseline for few-shot 3D point cloud classification. Furthermore, we propose a new network, Point-cloud Correlation Interaction (PCIA), with three novel plug-and-play components called Salient-Part Fusion (SPF) module, Self-Channel Interaction Plus (SCI+) module, and Cross-Instance Fusion Plus (CIF+) module to obtain more representative embeddings and improve the feature distinction. These modules can be inserted into most FSL algorithms with minor changes and significantly improve the performance. Experimental results on three benchmark datasets, ModelNet40-FS, ShapeNet70-FS, and ScanObjectNN-FS, demonstrate that our method achieves state-of-the-art performance for the 3D FSL task.
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