Moving Least-Squares Reconstruction of Large Models with GPUs

Modern laser range scanning campaigns produce extremely large point clouds, and reconstructing a triangulated surface thus requires both out-of-core techniques and significant computational power. We present a GPU-accelerated implementation of the moving least-squares (MLS) surface reconstruction technique. We believe this to be the first GPU-accelerated, out-of-core implementation of surface reconstruction that is suitable for laser range-scanned data. While several previous out-of-core approaches use a sweep-plane approach, we subdivide the space into cubic regions that are processed independently. This independence allows the algorithm to be parallelized using multiple GPUs, either in a single machine or a cluster. It also allows data sets with billions of point samples to be processed on a standard desktop PC. We show that our implementation is an order of magnitude faster than a CPU-based implementation when using a single GPU, and scales well to 8 GPUs.

[1]  Gabriel Taubin,et al.  The ball-pivoting algorithm for surface reconstruction , 1999, IEEE Transactions on Visualization and Computer Graphics.

[2]  Renato Pajarola,et al.  Fast low-memory streaming MLS reconstruction of point-sampled surfaces , 2009, Graphics Interface.

[3]  Nina Amenta,et al.  GPU-Assisted Surface Reconstruction on Locally-Uniform Samples , 2008, IMR.

[4]  Message P Forum,et al.  MPI: A Message-Passing Interface Standard , 1994 .

[5]  Markus H. Gross,et al.  Algebraic point set surfaces , 2007, ACM Trans. Graph..

[6]  Michael M. Kazhdan,et al.  Poisson surface reconstruction , 2006, SGP '06.

[7]  Andrew S. Grimshaw,et al.  Parallel Scan for Stream Architectures , 2012 .

[8]  Randal C. Burns,et al.  Parallel Poisson Surface Reconstruction , 2009, ISVC.

[9]  Message Passing Interface Forum.  MPI: A message - passing interface standard , 1994 .

[10]  Bernd Hamann,et al.  The asymptotic decider: resolving the ambiguity in marching cubes , 1991, Proceeding Visualization '91.

[11]  Gabriel Taubin,et al.  A benchmark for surface reconstruction , 2013, TOGS.

[12]  Paolo Cignoni,et al.  Out-of-core MLS reconstruction , 2007 .

[13]  Arthur W. Toga,et al.  Surface mapping brain function on 3D models , 1990, IEEE Computer Graphics and Applications.

[14]  Robert Sedgewick,et al.  Algorithms in C , 1990 .

[15]  Paolo Cignoni,et al.  Practical and Robust MLS-based Integration of Scanned Data , 2008, Eurographics Italian Chapter Conference.

[16]  Marc Alexa,et al.  On Normals and Projection Operators for Surfaces Defined by Point Sets , 2004, PBG.

[17]  Heinz Rüther,et al.  Challenges in Heritage Documentation with Terrestrial Laser Scanning , 2011 .

[18]  Randal C. Burns,et al.  Multilevel streaming for out-of-core surface reconstruction , 2007, Symposium on Geometry Processing.

[19]  Aaftab Munshi,et al.  The OpenCL specification , 2009, 2009 IEEE Hot Chips 21 Symposium (HCS).

[20]  Paolo Cignoni,et al.  MeshLab: an Open-Source Mesh Processing Tool , 2008, Eurographics Italian Chapter Conference.

[21]  Nadia Magnenat-Thalmann,et al.  Surface from Scattered Points: A Brief Survey of Recent Developments , 2005 .

[22]  A. Adamson,et al.  Approximating bounded, nonorientable surfaces from points , 2004, Proceedings Shape Modeling Applications, 2004..

[23]  Renato Pajarola,et al.  Flexible Configurable Stream Processing of Point Data , 2009 .

[24]  Marc Alexa,et al.  Approximating Bounded, Non-Orientable Surfaces from Points (Figures 5, 6, and 7) , 2004, SMI.

[25]  Marc Alexa,et al.  Point set surfaces , 2001, Proceedings Visualization, 2001. VIS '01..

[26]  A. Adamson,et al.  Ray tracing point set surfaces , 2003, 2003 Shape Modeling International..

[27]  Michael M. Kazhdan,et al.  Screened poisson surface reconstruction , 2013, TOGS.

[28]  Markus H. Gross,et al.  Dynamic Sampling and Rendering of Algebraic Point Set Surfaces , 2008, Comput. Graph. Forum.

[29]  Michael M. Kazhdan,et al.  Unconstrained isosurface extraction on arbitrary octrees , 2007, Symposium on Geometry Processing.

[30]  Kun Zhou,et al.  Data-Parallel Octrees for Surface Reconstruction , 2011, IEEE Transactions on Visualization and Computer Graphics.

[31]  James E. Gain,et al.  Fast in-place binning of laser range-scanned point sets , 2013, JOCCH.

[32]  Abel J. P. Gomes,et al.  CUDA-based triangulations of convolution molecular surfaces , 2010, HPDC '10.