Point Cloud to Mesh Reconstruction: Methods, Trade-offs, and Implementation Guide
Not on the map. Why?
Abstract (arXiv)
Reconstructing meshes from point clouds is a fundamental task in computer vision with applications spanning robotics, autonomous systems, and medical imaging. Selecting an appropriate learning-based method requires understanding trade-offs between computational efficiency, geometric accuracy, and output constraints. This paper categorizes over fifteen methods into five paradigms -- PointNet family, autoencoder architectures, deformation-based methods, point-move techniques, and primitive-based approaches -- and provides practical guidance for method selection. We contribute: (1) a decision framework mapping input/output requirements to suitable paradigms, (2) a failure mode analysis to assist practitioners in debugging implementations, (3) standardized comparisons on ShapeNet benchmarks, and (4) a curated list of maintained codebases with implementation resources. By synthesizing both theoretical foundations and practical considerations, this work serves as an entry point for practitioners and researchers new to learning-based 3D mesh reconstruction.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 29 of its 65 references that are in the map. First posted on arXiv 14 Dec 2024.
Builds on
29 of 65 references.
References · 29
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
- 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
- 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
- 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
- 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,682
- 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
- 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,037
- 2016Density estimation using Real NVPcited by 2,029
- 2019Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regressioncited by 1,361
- 2014NICE: Non-linear Independent Components Estimationcited by 1,052
- 2016A Point Set Generation Network for 3D Object Reconstruction from a Single Imagecited by 972
- 20163D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstructioncited by 587
17 more
- 2022PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategiescited by 432
- 2018PCN: Point Completion Networkcited by 382
- 2019BSP-Net: Generating Compact Meshes via Binary Space Partitioningcited by 205
- 2016Thingi10K: A Dataset of 10,000 3D-Printing Modelscited by 182
- 2018Deep Geometric Prior for Surface Reconstructioncited by 110
- 2015Benchmarking in Manipulation Research: The YCB Object and Model Set and Benchmarking Protocolscited by 98
- 2020ParSeNet: A Parametric Surface Fitting Network for 3D Point Cloudscited by 96
- 20193DN: 3D Deformation Networkcited by 60
- 2020Neural Mesh Flow: 3D Manifold Mesh Generation via Diffeomorphic Flowscited by 60
- 2020Generative PointNet: Deep Energy-Based Learning on Unordered Point Sets for 3D Generation, Reconstruction and Classificationcited by 57
- 2022CaDeX: Learning Canonical Deformation Coordinate Space for Dynamic Surface Representation via Neural Homeomorphismcited by 56
- 2020Meshing Point Clouds with Predicted Intrinsic-Extrinsic Ratio Guidancecited by 39
- 2021HPNet: Deep Primitive Segmentation Using Hybrid Representationscited by 35
- 2020PointTriNet: Learned Triangulation of 3D Point Setscited by 32
- 2021Learning Parallel Dense Correspondence from Spatio-Temporal Descriptors for Efficient and Robust 4D Reconstructioncited by 31
- 2023Fast Point Cloud to Mesh Reconstruction for Deformable Object Trackingcited by 4
- 2022Isomorphic mesh generation from point clouds with multilayer perceptronscited by 1
Built on by
Not cited yet.