Robust kernel-based feature representation for 3D point cloud analysis via circular convolutional networkopen access
- Authors
- Jung, S.; Shin, Y.-G.; Chung, M.
- Issue Date
- Jun-2023
- Publisher
- Academic Press Inc.
- Keywords
- 3D point cloud analysis; Angle-based kernel convolutions; Global context aggregation; Rotation-robust point descriptor; Scale adaptation
- Citation
- Computer Vision and Image Understanding, v.231
- Journal Title
- Computer Vision and Image Understanding
- Volume
- 231
- URI
- https://scholarworks.bwise.kr/ssu/handle/2018.sw.ssu/43911
- DOI
- 10.1016/j.cviu.2023.103678
- ISSN
- 1077-3142
1090-235X
- Abstract
- Feature descriptors of point clouds are used in several applications, such as registration and part segmentation of 3D point clouds. Learning representations of local geometric features is unquestionably the most important task for accurate point cloud analyses. However, it is challenging to develop rotation or scale-invariant descriptors. Most previous studies have either ignored rotations or empirically studied optimal scale parameters, which hinders the applicability of the methods for real-world datasets. In this paper, we present a new local feature description method that is robust to rotation and scale variations. Moreover, we improve representations based on a global aggregation method. First, we place kernels aligned around each point in the normal direction. To avoid the sign problem of the normal vector, we use a symmetric kernel point distribution in the tangential plane. From each kernel point, we first project the points from the spatial space to the feature space, which is robust to multiple scales and rotation, based on angles and distances. Subsequently, we perform convolutions by considering local kernel point structures and long-range global context, obtained by a global aggregation method. We experimented with our proposed descriptors on benchmark datasets (i.e., ModelNet40 and ShapeNetPart) to evaluate the performance of registration, classification, and part segmentation on 3D point clouds. Our method showed superior performances when compared to the state-of-the-art methods by reducing 70% of the rotation and translation errors in the registration task. Our method also showed comparable performance in the classification and part-segmentation tasks without any external data augmentation. © 2023 Elsevier Inc.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - College of Information Technology > School of Software > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.