Navigation

PetroSurf3D – A high-resolution 3D Dataset of Rock Art for Surface Segmentation*

Georg Poier1, Markus Seidl2, Matthias Zeppelzauer2, Christian Reinbacher1, Martin Schaich3, Giovanna Bellandi4, Alberto Marretta4, Horst Bischof1

1Institute for Computer Graphics and Vision, Graz University of Technology
2Institute of Creative\Media/Technologies, St. Pölten University of Applied Sciences
3ArcTron 3D
4McDonald Institute for Archaeological Research, University of Cambridge
5Parco Archeologico Comunale di Seradina-Bedolina

* The research leading to these results has been carried out as part of the project 3D-Pitoti (3d-pitoti.eu), which was funded from the European Community's Seventh Framework Programme (FP7/2007-2013) under grant agreement no. 600545; 2013-2016.





Example orthophotos (left), corresponding depth maps (center), and ground truth labels (right). For visualization of the depth we normalized and cropped the distance ranges per scan and show the resulting values in false color.
Example orthophoto Example depth map Example ground truth labelling
Example orthophoto Example depth map Example ground truth labelling
Example orthophoto Example depth map Example ground truth labelling

Motivation

Ancient rock engravings (so called petroglyphs) represent one of the earliest surviving artifacts describing life of our ancestors. Recently, modern 3D scanning techniques found their application in the domain of rock art documentation by providing high-resolution reconstructions of rock surfaces. Reconstruction results demonstrate the strengths of novel 3D techniques and have the potential to replace the traditional (manual) documentation techniques of archaeologists.

An important analysis task in rock art documentation is the segmentation of petroglyphs. To foster automation of this tedious step, we present a high-resolution 3D surface dataset of natural rock surfaces which exhibit different petroglyphs together with accurate expert ground-truth annotations. To our knowledge, this dataset is the first public 3D surface dataset which allows for surface segmentation at sub-millimeter scale. We conducted experiments with state-of-the-art methods to generate a baseline for the dataset and verified that the size and variability of the data is sufficient to successfully adopt even recent data-hungry Convolutional Neural Networks (CNNs). Furthermore, we experimentally demonstrated that the provided geometric information is key to successful automatic segmentation and strongly outperforms color-based segmentation. The introduced dataset represents a novel benchmark for 3D surface segmentation methods in general and is intended to foster comparability among different approaches in future.

Material

References

  1. PetroSurf3D - A high-resolution 3D Dataset of Rock Art for Surface Segmentation (bib) (project)Georg Poier, Markus Seidl, Matthias Zeppelzauer, Christian Reinbacher, Martin Schaich, Giovanna Bellandi, Alberto Marretta, Horst Bischof CoRR abs/1610.01944, 2016
  1. Interactive 3D Segmentation of Rock-Art by Enhanced Depth Maps and Gradient Preserving Regularization (bib)Matthias Zeppelzauer, Georg Poier, Markus Seidl, Christian Reinbacher, Samuel Schulter, Christian Breiteneder, and Horst Bischof ACM Journal on Computing and Cultural Heritage 9 (4): 19:1-19:30, 2016
    (The original publication is available at http://dx.doi.org/10.1145/2950062)
  1. Interactive Segmentation of Rock-Art in High-Resolution 3D Reconstructions (bib)Matthias Zeppelzauer, Georg Poier, Markus Seidl, Christian Reinbacher, Christian Breiteneder, Horst Bischof, and Samuel Schulter In Proc. Digital Heritage Conference (DH), 2015
    (oral presentation; winner of the best paper price)

Copyright 2010 ICG

Valid XHTML 1.1 Valid CSS