Abstract:In this work, we study the perception problem for sampled surfaces (possibly with boundary) using tools from computational topology, specifically, how to identify their underlying topology starting from point-cloud samples in space, such as those obtained with 3D scanners. We present a reconstruction algorithm based on a careful topological study of the point sample that allows us to obtain a cellular decomposition of it using a Morse function. No triangulation or local implicit equations are used as intermediate steps, avoiding in this way reconstruction-induced artifices. The algorithm can be run without any prior knowledge of the surface topology, density or regularity of the point-sample. The results consist of a piece-wise decomposition of the given surface as a union of Morse cells (i.e. topological disks), suitable for tasks such as mesh-independent reparametrization or noise-filtering, and a small-rank cellular complex determining the topology of the surface. The algorithm, which we test with several real and synthetic surfaces, can be applied to smooth surfaces with or without boundary, embedded in an ambient space of any dimension.

What problem does this paper attempt to address?

The problem this paper attempts to address is: how to identify the fundamental topological structure of sampled surfaces (possibly with boundaries) from point cloud samples in space (e.g., point clouds obtained through 3D scanners). Specifically, the paper proposes a reconstruction algorithm based on Morse theory that can directly handle point cloud data to generate a cellular decomposition of the surface without the need for intermediate triangulation or local implicit equation calculations. This algorithm is capable of preserving the topological features of the surface without prior knowledge and is applicable to smooth or boundary surfaces. ### Main Issues: 1. **Identifying Surface Topology**: How to identify the basic topological structure of the surface from point cloud samples, including its connectivity, holes, and other features. 2. **Avoiding Reconstruction Artifacts**: Generating a cellular decomposition of the surface directly from the point cloud without introducing intermediate steps (such as triangulation or local implicit equations), thereby avoiding potential artifacts during reconstruction. 3. **Handling Surfaces with Boundaries**: Many existing surface reconstruction algorithms primarily focus on closed surfaces without boundaries, while this paper proposes methods capable of handling surfaces with boundaries. ### Solution: - **Morse Theory**: Utilizing Morse functions and their gradient flows to identify critical points (such as minima, saddles, and maxima) in the point cloud and generating Morse cells based on these critical points. - **Cellular Decomposition**: Generating a cellular decomposition of the surface through the connectivity of Morse cells, thereby determining the topological structure of the surface. - **Boundary Detection**: Developing a graph-theoretic method to robustly detect boundary points in the point cloud, ensuring that surfaces with boundaries can be correctly reconstructed. ### Application Scenarios: - **Robotic Manipulation**: In robotic manipulation of textiles, it is necessary to reconstruct and identify the topology of clothing from point cloud data for effective operation. - **Surface Reconstruction**: In computer graphics and computational geometry, reconstructing surfaces from point cloud data is an important problem, and this algorithm can be applied to various surface reconstruction tasks. ### Innovations: - **Direct Handling of Point Clouds**: No need for intermediate triangulation or local implicit equation calculations, directly generating cellular decomposition from point clouds. - **Robustness**: Capable of preserving the topological features of the surface without prior knowledge, suitable for surfaces of various complex shapes. - **Handling Surfaces with Boundaries**: Special focus on surfaces with boundaries, developing new methods to detect boundary points, ensuring the accuracy of reconstruction results. In summary, this paper addresses the problem of reconstructing the topological structure of surfaces with boundaries from point cloud data through Morse theory and cellular decomposition methods, with broad application prospects.

Topological reconstruction of sampled surfaces via Morse theory

Reconstruction of sampled surfaces with boundary via Morse theory

Determining the Topology of Real Algebraic Surfaces

A new research of surfaces reconstruction of revolution in reverse engineering

An Algorithm for Topology Reconstruction from Unorganized Points Based on Local Flatness of Surface

Point Sampled Surface Reconstruction Based On Local Geometry

Topological Skeletonization and Tree-Summarization of Neurons Using Discrete Morse Theory

Intelligent sampling for the measurement of structured surfaces

Topology-controllable Implicit Surface Reconstruction Based on Persistent Homology

Incremental Topological Surface Reconstruction from Massive Point Cloud Data

Applications of Forman's discrete Morse theory to topology visualization and mesh compression

Multi-scale Surface Reconstruction Based on a Curvature-Adaptive Signed Distance Field.

Morse Theory for Chromatic Delaunay Triangulations

Feature Correspondences Using Morse Smale Complex

NURBS-based surface generation from 3D images: spectral construction and data-driven model selection

Distortion-Free Intelligent Sampling of Sparse Surfaces Via Locally Refined T-Spline Metamodelling

Morse Theoretic Signal Compression and Reconstruction on Chain Complexes

Describing shapes by geometrical-topological properties of real functions

Reconstructing Curves from Sparse Samples on Riemannian Manifolds

Topology Reconstruction of BIM Wall Objects from Point Cloud Data

A Survey of Surface Reconstruction from Point Clouds