arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2501.07076cs.CVcs.AI
收藏

全局和局部输入下的点云上采样表示学习

Representation Learning of Point Cloud Upsampling in Global and Local Inputs

East China University of Science and Technology(东华大学);The Hong Kong Polytechnic University(香港理工大学)

Tongxu Zhang, Bei Wang

AI总结:

本研究提出ReLPU框架,通过结合全局和局部特征学习提升点云上采样的性能和可解释性。

摘要:

近年来,点云上采样已被广泛应用于3D重建和物体识别等任务。本研究提出了一种新的框架ReLPU,通过显式学习点云的全局和局部结构特征来提升上采样性能。具体而言,我们从均匀分割的输入中提取全局特征(平均段),并从相同点云的基于块的输入中提取局部特征。这两种特征通过并行自编码器处理,融合后送入共享解码器进行上采样。双输入设计提高了特征完整性和跨尺度一致性,特别是在稀疏和噪声区域。我们的框架应用于几种最先进的自编码器网络,并在标准数据集上进行了验证。实验结果表明,在几何保真度和鲁棒性方面均取得了持续改进。此外,显著性图证实了平行全局-局部学习显著增强了点云上采样的可解释性和性能。

英文摘要:

In recent years, point cloud upsampling has been widely applied in tasks such as 3D reconstruction and object recognition. This study proposed a novel framework, ReLPU, which enhances upsampling performance by explicitly learning from both global and local structural features of point clouds. Specifically, we extracted global features from uniformly segmented inputs (Average Segments) and local features from patch-based inputs of the same point cloud. These two types of features were processed through parallel autoencoders, fused, and then fed into a shared decoder for upsampling. This dual-input design improved feature completeness and cross-scale consistency, especially in sparse and noisy regions. Our framework was applied to several state-of-the-art autoencoder-based networks and validated on standard datasets. Experimental results demonstrated consistent improvements in geometric fidelity and robustness. In addition, saliency maps confirmed that parallel global-local learning significantly enhanced the interpretability and performance of point cloud upsampling.

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。

↑