Version 1
: Received: 3 May 2023 / Approved: 6 May 2023 / Online: 6 May 2023 (05:19:09 CEST)
How to cite:
Li, M.; Zhao, Y.; Zhou, Y.; Wei, J. Point-MDA: Progressive Point Cloud Analysis for Extreme Multi-Scale Detail Activation. Preprints2023, 2023050379. https://doi.org/10.20944/preprints202305.0379.v1
Li, M.; Zhao, Y.; Zhou, Y.; Wei, J. Point-MDA: Progressive Point Cloud Analysis for Extreme Multi-Scale Detail Activation. Preprints 2023, 2023050379. https://doi.org/10.20944/preprints202305.0379.v1
Li, M.; Zhao, Y.; Zhou, Y.; Wei, J. Point-MDA: Progressive Point Cloud Analysis for Extreme Multi-Scale Detail Activation. Preprints2023, 2023050379. https://doi.org/10.20944/preprints202305.0379.v1
APA Style
Li, M., Zhao, Y., Zhou, Y., & Wei, J. (2023). Point-MDA: Progressive Point Cloud Analysis for Extreme Multi-Scale Detail Activation. Preprints. https://doi.org/10.20944/preprints202305.0379.v1
Chicago/Turabian Style
Li, M., Yun Zhou and Jiang Wei. 2023 "Point-MDA: Progressive Point Cloud Analysis for Extreme Multi-Scale Detail Activation" Preprints. https://doi.org/10.20944/preprints202305.0379.v1
Abstract
The point cloud is a form of three-dimensional data that comprises various detailed features at multiple scales. Due to this characteristic and its irregularity, point cloud analysis based on deep learning is challenging. While previous works utilize the sampling-grouping operation of PointNet++ for feature description and then explore geometry by means of sophisticated feature extractors or deep networks, such operations fail to describe multi-scale features effectively. Additionally, these techniques have led to performance saturation. And it is difficult for standard MLPs to directly "mine" point cloud. To address above problems, we propose the Detail Activation (DA) module, which encodes data based on Fourier transform after sampling and grouping. We activate the channels at different frequency levels from low to high in the DA module to gradually recover finer point cloud details. As training progresses, the proposed Point-MDA can uncover local and global geometries of point cloud progressively. Our experiments show that Point-MDA achieves superior classification accuracy, outperforming PointNet++ by 3.3% and 7.9% in terms of overall accuracy on the ModelNet40 and ScanObjectNN dataset, respectively. Furthermore, it accomplishes this without employing complicated operations, while exploring the full potential of PointNet++.
Keywords
point cloud classification; deep learning; computer vision; scene understanding
Subject
Computer Science and Mathematics, Computer Vision and Graphics
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
