SDFConnect: Neural Implicit Surface Reconstruction of a Sparse Point Cloud with Topological Constraints

We present a novel approach for neural implicit surface reconstruction from relatively sparse point cloud to ensure the reconstruction of a single connected component. We introduce a topological loss term based on persistent homology to reconstruct a manifold object of genus 1. Building on the Neura...

Full description

Saved in:
Bibliographic Details
Main Authors: Jignasu, Anushrut, Balu, Aditya, Sarkar, Soumik, Hegde, Chinmay, Ganapathysubramanian, Baskar, Krishnamurthy, Adarsh
Format: Conference Proceeding
Language:English
Subjects:
Geometry
Manifolds
persistent homology
Pipelines
Point cloud compression
Shape
signed distance fields
Surface reconstruction
Three-dimensional displays
topological data analysis
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present a novel approach for neural implicit surface reconstruction from relatively sparse point cloud to ensure the reconstruction of a single connected component. We introduce a topological loss term based on persistent homology to reconstruct a manifold object of genus 1. Building on the Neural Pull [25] framework, our method demonstrates superior performance in preserving the integrity of complex 3D geometries, evident through both visual and empirical comparisons. Our contributions include the integration of persistent diagrams to refine shape topology and a topological loss term to constrain existing reconstruction pipelines to a single connected component. This advancement allows for the seamless integration of topological data analysis with implicit surface reconstruction.
ISSN:2160-7516
DOI:10.1109/CVPRW63382.2024.00536