计算机科学
离群值
点云
人工智能
模式识别(心理学)
联营
特征提取
稳健性(进化)
加权
数据挖掘
医学
生物化学
化学
放射科
基因
作者
Yuxin Zhang,Zhan-Li Sun,Zhigang Zeng,Kin-Man Lam
出处
期刊:IEEE transactions on artificial intelligence
[Institute of Electrical and Electronics Engineers]
日期:2023-10-01
卷期号:4 (5): 1317-1327
标识
DOI:10.1109/tai.2022.3201505
摘要
How to accurately register partial point cloud still remains a challenging task, because of its irregular and unordered structure in a non-Euclidean space, noise, outliers, and other unfavorable factors. In this article, an effective partial point cloud registration network is proposed by devising a two-stage deep local feature extraction process and an outlier filtering strategy. To be specific, on the one hand, to effectively capture geometric interdependency in the low-level space, a local attention feature extraction module is explored to extract local contextual attention features by highlighting different attention weights on neighborhoods. On the other hand, in the local feature aggregation module, two position encoding blocks are applied to increase the receptive field of each point in the high-level space. Of these, an attentive pooling can automatically learn important local features to alleviate the possible information loss. Furthermore, to derive the weight of the putative correspondence, an outlier filtering module is designed by consisting of point context normalization block, differentiable pooling layer, and differentiable unpooling layer. Moreover, in order to enhance robustness, a weighting point cloud registration model is formulated to alleviate outliers by considering the contribution of each correspondence. Experiments on multiple datasets demonstrate that the proposed approach is competitive to several state-of-the-art algorithms.
科研通智能强力驱动
Strongly Powered by AbleSci AI