arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2209.11796cs.CVcs.LG
收藏

Composite Convolution:一种用于3D点云深度学习的灵活算子

Composite Convolution: a Flexible Operator for Deep Learning on 3D Point Clouds

Politecnico di Milano(米兰理工大学);STMicroelectronics(意法半导体)

Alberto Floris, Luca Frittoli, Diego Carrera, Giacomo Boracchi

AI总结:

提出了一种用于3D点云深度学习的灵活算子composite layer,通过提取压缩空间信息并与特征向量结合实现网络设计的灵活性和正则化,基于此构建的CompositeNets在分类、分割和异常检测任务中表现优异。

摘要:

深度神经网络需要特定的层来处理点云,因为3D点分散且不规则的位置阻碍了传统卷积滤波器的使用。我们引入了composite layer,这是一种用于处理3D点云的现有卷积算子的灵活且通用的替代方案。我们将composite layer设计为从点的3D坐标中提取并压缩空间信息,然后将其与特征向量相结合。与ConvPoint和KPConv等主流点卷积层相比,我们的composite layer保证了网络设计中的更大灵活性,并提供了额外的正则化形式。为了展示我们composite layers的通用性,我们定义了一个卷积composite layer和一个以非线性方式结合空间信息和特征的聚合版本,并使用这些层来实现CompositeNets。我们在合成和真实世界数据集上的实验表明,在分类、分割和异常检测中,我们的CompositeNets优于使用相同序列架构的ConvPoint,并取得了与具有更深残差架构的KPConv相似的结果。此外,我们的CompositeNets在点云异常检测方面达到了最先进的性能。我们的代码在\url{https://github.com/sirolf-otrebla/CompositeNet}上公开提供。

英文摘要:

Deep neural networks require specific layers to process point clouds, as the scattered and irregular location of 3D points prevents the use of conventional convolutional filters. We introduce the composite layer, a flexible and general alternative to the existing convolutional operators that process 3D point clouds. We design our composite layer to extract and compress the spatial information from the 3D coordinates of points and then combine this with the feature vectors. Compared to mainstream point-convolutional layers such as ConvPoint and KPConv, our composite layer guarantees greater flexibility in network design and provides an additional form of regularization. To demonstrate the generality of our composite layers, we define both a convolutional composite layer and an aggregate version that combines spatial information and features in a nonlinear manner, and we use these layers to implement CompositeNets. Our experiments on synthetic and real-world datasets show that, in both classification, segmentation, and anomaly detection, our CompositeNets outperform ConvPoint, which uses the same sequential architecture, and achieve similar results as KPConv, which has a deeper, residual architecture. Moreover, our CompositeNets achieve state-of-the-art performance in anomaly detection on point clouds. Our code is publicly available at \url{https://github.com/sirolf-otrebla/CompositeNet}.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑