Distributed Poisson Surface Reconstruction

M. Kazhdan; H. Hoppe

doi:10.1111/cgf.14925

journal article Aug 31, 2023

Distributed Poisson Surface Reconstruction

M. Kazhdan

H. Hoppe

Computer Graphics Forum Vol. 42 No. 6 · Wiley

View at Publisher Save 10.1111/cgf.14925

Abstract

AbstractScreened Poisson surface reconstruction robustly creates meshes from oriented point sets. For large datasets, the technique requires hours of computation and significant memory. We present a method to parallelize and distribute this computation over multiple commodity client nodes. The method partitions space on one axis into adaptively sized slabs containing balanced subsets of points. Because the Poisson formulation involves a global system, the challenge is to maintain seamless consistency at the slab boundaries and obtain a reconstruction that is indistinguishable from the serial result. To this end, we express the reconstructed indicator function as a sum of a low‐resolution term computed on a server and high‐resolution terms computed on distributed clients. Using a client–server architecture, we map the computation onto a sequence of serial server tasks and parallel client tasks, separated by synchronization barriers. This architecture also enables low‐memory evaluation on a single computer, albeit without speedup. We demonstrate a 700 million vertex reconstruction of the billion point David statue scan in less than 20 min on a 65‐node cluster with a maximum memory usage of 45 GB/node, or in 14 h on a single node.

Topics

No keywords indexed for this article. Browse by subject →

References

36

[1]

[ABCO*01] AlexaM. BehrJ. Cohen‐OrD. FleishmanS. LevinD. SilvaC.:Point set surfaces. InProceedings of the Conference on Visualization '01(2001).

[2]

A new Voronoi-based surface reconstruction algorithm

Nina Amenta, Marshall Bern, Manolis Kamvysselis

Proceedings of the 25th annual conference on Compu... 10.1145/280814.280947

[3]

10.1109/tvcg.2006.169

[4]

[BGKS20] BadkiA. GalloO. KautzJ. SenP.:Meshlet priors for 3D mesh reconstruction. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition(2020) pp.2849–2858. 10.1109/cvpr42600.2020.00292

[5]

[BKBH07] BolithoM. KazhdanM. BurnsR. HoppeH.:Multilevel streaming for out‐of‐core surface reconstruction. InSymposium on Geometry Processing(2007) pp.69–78.

[6]

10.1007/978-3-642-10331-5_63

[7]

10.1109/2945.817351

[8]

10.1145/2461912.2461994

[9]

[CL96a] CurlessB. LevoyM.:A volumetric method for building complex models from range images. InComputer Graphics (Proceedings of SIGGRAPH '96)(1996). 10.1145/237170.237269

[10]

[CL96b] CurlessB. LevoyM.:A volumetric method for building complex models from range images. InProceedings of the 23rd Annual Conference on Computer Graphics and Interactive Techniques(1996) pp.303–312. 10.1145/237170.237269

[11]

10.1111/j.1467-8659.2011.02058.x

[12]

Chen Z. "Neural dual contouring" ACM Transactions on Graphics (TOG) (2022) 10.1145/3528223.3530108

[13]

[DMP] The Digital Michelangelo Project.https://graphics.stanford.edu/papers/digmich_falletti/

[14]

10.1145/2601097.2601163

[15]

10.1109/3dv.2015.36

[16]

Surface reconstruction from unorganized points

Hugues Hoppe, Tony DeRose, Tom Duchamp et al.

ACM SIGGRAPH Computer Graphics 10.1145/142920.134011

[17]

10.1145/882262.882366

[18]

[IL05] IsenburgM. LindstromP.:Streaming meshes. InProceedings of the IEEE Visualization(2005) IEEE pp.231–238. 10.1109/visual.2005.1532800

[19]

[ILS05] IsenburgM. LindstromP. SnoeyinkJ.:Streaming compression of triangle meshes. InProceedings of the Symposium on Geometry Processing(2005). 10.1145/1187112.1187276

[20]

10.1145/1141911.1141992

[21]

[KBH06] KazhdanM. BolithoM. HoppeH.:Poisson surface reconstruction. InProceedings of the Symposium on Geometry Processing(2006) pp.61–70.

[22]

10.1145/2487228.2487237

[23]

[KKDH07] KazhdanM. KleinA. DalalK. HoppeH.:Unconstrained isosurface extraction on arbitrary octrees. InProceedings of the Symposium on Geometry Processing(2007) pp.125–133.

[24]

[LPC*00] LevoyM. PulliK. CurlessB. RusinkiewiczS. KollerD. PereiraL. GinztonM. AndersonS. DavisJ. GinsbergJ. ShadeJ. FulkD.:The digital Michelangelo project: 3D scanning of large statues. InProceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques(2000) pp.131–144. 10.1145/344779.344849

[25]

10.1145/3592440

[26]

10.1145/1289603.1289607

[27]

10.1145/2508363.2508374

[28]

10.1145/882262.882293

[29]

[Paj05] PajarolaR.:Stream‐processing points. InProceedings of the Conference on Visualization '05(2005).

[30]

[RGA*21] RakotosaonaM.‐J. GuerreroP. AigermanN. MitraN. J. OvsjanikovM.:Learning Delaunay surface elements for mesh reconstruction. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition(2021) pp.22–31. 10.1109/cvpr46437.2021.00009

[31]

[Resonai] https://www.resonai.com/

[32]

10.1145/3072959.3073635

[33]

[UB15] UmmenhoferB. BroxT.:Global dense multiscale reconstruction for a billion points. InProceedings of the IEEE International Conference on Computer Vision(2015) pp.1341–1349. 10.1109/iccv.2015.158

[34]

10.1109/tvcg.2007.21

[35]

[WSS*19] WilliamsF. SchneiderT. SilvaC. ZorinD. BrunaJ. PanozzoD.:Deep geometric prior for surface reconstruction. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition(2019) pp.10130–10139. 10.1109/cvpr.2019.01037

[36]

[ZGHG08] ZhouK. GongM. HuangX. GuoB.:Highly parallel surface reconstruction. Tech. Rep. Microsoft Research Asia 2008.

Metrics

15

Citations

36

References

Details