Abstract:3D shape analysis has drawn broad attention due to its increasing demands in various fields. Despite that impressive performance has been achieved on several databases, most researchers focus their ef-forts on improving the performance of shape classification, retrieval, segmentation, etc. They neglect the fact that the disturbances, such as orientation and deformation, may impact much on the perception, restricting the capacity of generalizing to real applications where the prior of orientation and pose is of-ten unknown. In this paper, we conduct shape analysis on point clouds and propose the point projection feature, which is rotation-invariant. Specifically, a novel architecture is designed to mine features of dif-ferent levels. We adopt a PointNet-based backbone to extract global feature for the point cloud, and the graph aggregation operation to perceive local pose variance in the Euclidean space or geodesic space. An efficient key point descriptor is designed to assign each point with different response and help recognize the overall geometry. Furthermore, a novel dataset, PKUnon-rigid, is built that is composed of non-rigid 3D objects, based on which we evaluate the capacity of several mainstream methods in terms of pro-cessing non-rigid shapes. Mathematical analyses and experimental results demonstrate that the proposed method can extract isometry-invariant representations for 3D shape analysis tasks without rotation aug-mentation, and outperforms other state-of-the-art methods. The proposed dataset is publicly available at https://github.com/tasx0823/PKUnon-rigid.(c) 2022 Elsevier Ltd. All rights reserved.

Shape Back-Projection In 3D Scenes

A Metric Learning Method for Image-based 3D Shape Retrieval

A Unified Feature Representation and Learning Framework for 3D Shape

Learning isometry-invariant representations for point cloud analysis

Surface Extraction and Boundary Detection Based on DBSCAN Clustering in 3D Point Clouds

Shape Representation and Recognition in High Dimensional Feature Space

Shape-aware Surface Reconstruction from Sparse 3D Point-Clouds

PC2: Projection-Conditioned Point Cloud Diffusion for Single-Image 3D Reconstruction

Back to 3D: Few-Shot 3D Keypoint Detection with Back-Projected 2D Features

From Points to Multi-Object 3D Reconstruction

RealPoint3D: Point Cloud Generation from a Single Image with Complex Background

Information Geometric Framework For Point Cloud Data

Projective Analysis for 3D Shape Segmentation

Bayesian Reconstruction of 3d Shapes and Scenes from A Single Image

$PC^2$: Projection-Conditioned Point Cloud Diffusion for Single-Image 3D Reconstruction

PCPNET: Learning Local Shape Properties from Raw Point Clouds

Hyperbolic-constraint Point Cloud Reconstruction from Single RGB-D Images

Nonparametric Regression for 3D Point Cloud Learning

Arbitrary point cloud upsampling via Dual Back-Projection Network

Primitive-based Shape Abstraction via Nonparametric Bayesian Inference

A Probabilistic Approach for Stereo 3D Point Cloud Reconstruction from Airborne Single-Channel Multi-Aspect SAR Image Sequences