:: A 3D Pottery Content Based Retrieval Benchmark Dataset ::

About the 3D Pottery Benchmark

The existence of benchmark data sets is essential in order to perform the evaluation and the comparison of content based retrieval methods. Although, no standard benchmark is available for 3D data shape matching there is currently a number of 3D data sets have been proposed by several research teams and are freely offered to be used.In order to evaluate our pottery specific 3D shape descriptors, we had to create a dataset of polygonal 3D vessels. The current pottery dataset is composed by a total number of 1012 digitised, manually modelled and semiautomated generated 3D models (3D vessel random generator). The content of the dataset is classified in several generic vessel shape categories such as Ancient Greek (Alabastron, Amphora, Hydria, Kantharos, Lekythos, Psykter, etc), Native American (Jar, Effigy, Bowl, Bottle, etc), modern pottery and others. The current classification has been performed with the help of the Department of Cultural Heritage of the ILSP/Athena Research Centre at an archaeologist-oriented semantic level. Please note that the current classification isn't the final one and we will be soon presenting a shape-oriented one. Furthermore, the 3D models are stored in the Wavefront OBJ file format and are currently followed by a single view thumbnail in JPEG file format and no texture information. A text file and an excel file is also provided where the filename of each model is assigned to a class. This information can be used for the computation of any performance metrics.

Content Providers

The following table gathers the content providers (authors/affiliation/references) of all the 3D models included in the current version of the dataset.


Content Provider

Number of Models


Downloaded from the 3D Millennium Website - http://www.m3corp.com/



Downloaded from the 3D Princeton Shape Benchmark Database - http://shape.cs.princeton.edu/benchmark/



Created with software described in: A. Koutsoudis, G. Pavlidis, F. Arnaoutoglou, D. Tsiafakis, C. Chamzas, qp: A Tool for Generating 3D Models of Ancient Greek Pottery, Journal of Cultural Heritage Vol. 10 (2009), pp. 281-295.



Downloaded from the Aim@Shape Website - http://www.aimatshape.net



Downloaded from the Archive 3D Website - http://archive3d.net/



Downloaded from the Blender 3D Model Repository



The 3D models were created with image-based 3D reconstruction method described in: Carlos Hernandez Esteban,Francis Schmitt,Silhouette and Stereo Fusion for 3D Object Modeling,Signal and Image Processing Department, CNRS UMR 5141 Ecole Nationale Superieure des Telecommunications, France,http://carlos-hernandez.org/3dmodeling.html. This work has been partially supported by the SCULPTEUR European project IST-2001-35372. We thank the Thomas Henry museum at Cherbourg for the image sequences corresponding to Oceania and Thomas Henry. http://www.tsi.enst.fr/3dmodels



The 3D models were digitised by the Clepsydra Digitisation Centre (http://clepsydra.ceti.gr), Athena Research Centre



Downloaded from the FileForestNarod.ru Website



The 3D models were provided by: Fraunhofer-Institute for Ceramic Technologies and Systems - http://www.ikts.fraunhofer.de/index_en.html



Downloaded from the Google Sketch Repository



The 3D models were manualy modelled and were used in: Horn-yeu Shiaw, Robert J.K. Jacob, Gregory R. Crane, The 3D vase museum: a new approach to context in a digital library,JCDL’04, June 7-11, 2004, Tucson, Arizona, USA. The 3D Vaze Museum - Robert Jacob (jacob@cs.tufts.edu),Horn-Yeu Shiaw (hshiaw@eecs.tufts.edu), Department of EECS, Tufts University, Medford, U.S.A.



Modelled by Anestis Koutsoudis (akoutsou@ipet.gr) using Blender. Initial images and texture maps were found in Wikipedia under the Louvre Museum thread



Downloaded from the NTU 3D Model Database - http://3d.csie.ntu.edu.tw



Provided by the Saxony’s Archaeological Heritage Service,http://sachsen.de/en/1841.htm,Dipl.-Ing. (FH) Thomas Reuter,Software-guided 3D-Documentation of Archaeological Finds,



Downloaded from the Sweethome3D - http://www.sweethome3d.eu



The models were provided by the Virtual Hampson Museum Project 2007-2008 (http://hampson.cast.uark.edu/),Hampson Archaeological Museum State park in Wilson,Arkansas,Dr. Fred Limp - CAST Director and Principal Investigator (Center’s Virtual Hampson Museum Development Team), The data are licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License. This dataset version DOES NOT include the complete data collection available on the Virtual Hampson Project website as it is pottery oriented and when we collected the data it was still a work in progress.


More details about the content providers and the models they have contributed in the dataset can be found in this Microsoft Excel file

The 3D Pottery Dataset

i. Download the complete 3D pottery dataset (version 1), including thumbnail images and groundtruth-metadata files --> 3D Pottery Benchmark Dataset Ver. 1

ii. Download thumbnail images only --> Thumbnails Ver. 1

iii. Single image thumbnails index of all the 3D models in the dataset --> Single Image Thumbnail Index

vi. Single image thumbnails index per vessel class --> Abstract, Amphora, Alabastron, Aryballos, Bowl, Dinos, Hydria, Kalathos, Kantharos, Krater, Kyathos, Kylix, Lagynos, Lebes, Lekythos, Lydion, Mastos, Modern-Bottle, Modern-Glass, Modern-Mug, Modern-Vase, Mug, Native American - Bottle, Native American - Bowl, Native American - Effigy, Native American - Jar, Nestoris, Oinochoe, Other, Pelike, Picher Shaped, Pithoeidi, Pithos, Psykter, Pyxis, Skyphos,

The 3D Pottery Version 1 interclass shape variation

Performing Retrieval Tests

For implementing retrieval experiments and shape descriptor performance evaluation we have used five generic shape categories (Ancient Greek Alabastron, Amphora, Psykter, Lekythos and Hydria) due to their relatively large population in the dataset and geometrical coherence. Other classes found in the dataset can also be used. No part of the dataset has been defined as training set. In our experiments we have used all models of a class as queries against the whole dataset. The current dataset comes with an initial classification that is based on some generic shape categories. Thus, although in terms of semantics the 3D models do belong to the same class they don't always depict similar geometrical properties.

Citation-Referencing of the 3D pottery benchmark dataset

Please reference to the following paper: A. Koutsoudis et al., 3D Pottery Content Based Retrieval based on Pose Normalisation and Segmentation, Journal of Cultural Heritage, Vol. 11 (2010), pp. 329-338.


The 3D pottery benchmark dataset is AVAILABLE ONLY for research purposes and to promote 3D content based retrieval and 3D data analysis in the cultural heritage domain. If you would like to revise,extend,recompile the current dataset please email the initial author. The 3D models offered by the Virtual Hampson Museum project are licensed under a Creative Commons Attribution-Non-commercial-Share Alike 3.0 Unported License. When you use specific models from this repository please make sure to also acknowledge the original source of the models and also to comply with their IPR terms and restrictions.


Please let us know if you have used the 3D pottery benchmark dataset for your research or you would like to contribute in enhancing the dataset by providing additional 3D models.


Anestis Koutsoudis BSc, MSc, Phd Computer Science
Associate Research Fellow - Multimedia Department
Institute for Language and Speech Processing - Research Centre 'Athena'
PO BOX 159, Kimmeria Campus, 67100, Xanthi, Greece
Tel. +30 2541 0 78787 (326) Fax. +30 2541 0 63656 Personal Website

Other Related Websites

Experimental Content-based 3D Pottery Search Engine
3D Scene Search Engine
Experimental content based retrieval engine in a real time 3D environment
Qp - A 3D Vessel Modeller - Random Generator


This work is part of the 03ED679 research project, implemented within the framework of the “Reinforcement Programme of Human Research Manpower” (PENED) and co-financed by National and Community Funds (25% from the Greek Ministry of Development – General Secretariat of Research and Technology and 75% from EU – European Social Fund). The authors would like to acknowledge and thank the Cultural Heritage Unit of the ILSP/R.C. ‘Athena’ for their support in this work, all content providers and more specifically Horn-yeu Shiaw, Robert Jacob and Gregory R. Crane from Tufts University, Tom Funkhouser from Princeton University, Fred Limp from the Centre of Advanced Spatial Technologies of the Hampson Museum, Carlos Hernández Esteban and Francis Schmitt from Télécom Paris for allowing the use of their 3D content.

Online at April 2013