A Benchmark for Surface Reconstruction


Surface reconstruction is a well studied problem, with a substantial amount of work done over the past two decades. However, a comprehensive means of evaluating surface reconstruction algorithms is noticably absent. This project is focused on deriving a methodology for the evaluation of surface reconstruction. This website accompanies our paper by providing all of the associated data and software. We hope that surface reconstruction practitioners will find the data useful for evaluating their own reconstruction algorithms, and furthering the development of the field.


Matthew Berger, Joshua A. Levine, Luis Gustavo Nonato, Gabriel Taubin, and Claudio T. Silva
A Benchmark for Surface Reconstruction
ACM Transactions on Graphics 32(2), 20:1-20:17 (to be presented at SIGGRAPH 2013)
ACM COPYRIGHT NOTICE. Copyright © 2013 by the Association for Computing Machinery, Inc. Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Publications Dept., ACM, Inc., fax +1 (212) 869-0481, or permissions@acm.org.

Point Clouds

We provide renderings, in the form of splats, of all of our point clouds for our error distribution experiments. Additionally, for each point cloud we also provide an accompanying movie of the laser-striped range scans performed.

Reconstruction Results

For each point cloud in our error distribution experiments, we show the meshes produced by all reconstruction algorithms. Additionally, we also plot the different error metrics, so one can obtain an immediate comparison across all algorithms.


We have setup a sourceforge project for our benchmark. The source code may be obtained through svn. Please see the README file at the top-level directory for further information.


Please send an email to us if you have any questions.