SHREC'17 track: Retrieval of surfaces with similar relief patterns

The aim of this SHREC'17 track is to evaluate the performance of retrieval algorithms when characterized by different relief patterns, such as knitted fabrics, etc.. The track is jointly organized by IMATI CNR and University of Verona. The SHREC17 track: Retrieval of surfaces with similar relief patterns final report will appear, after revision, in the proceedings of the EUROGRAPHICS Workshop on 3D Object Retrieval (3DOR).

Dataset, ground truth and evaluation
A set of examples is made available. Such a set is a sample of some patterns in different poses with four different samplings on the mesh. Intuitively, objects with the same "name" belong to the same class. Please click here to download the of examples that can be used to test the methods and tune their parameters.

The dataset consists of a 720 surfaces, grouped in 15 classed of 48 elements each one. Each class represents a single pattern. Each class is created by acquiring a single pattern in different poses and applying three processing operations (two subsamplings and one simplification) to each scan. Transformations have been designed to alter the connectivity of the original scan. The models will be randomly renamed. Since data are acquired with a depth sensor there might be small artefacts and small topological handles. Each surface is oriented, a single connected component, locally a regular, i.e. without dangling edges.

Each model will be used in turn as a query against the remaining part of the database. For a given query, the goal is to retrieve the most similar objects. The relevance or non-relevance of the models for a given query, i.e. the ground truth, will be established a priori. The performance of the retrieval algorithms will be evaluated using the precision-recall curves, nearest neighbor, first tier, second tier, normalized discounted cumulated gain and average dynamic recall.

Registration and other procedures
Each participant is requested to register to the track by sending an email to Silvia Biasotti (email: silvia.biasotti@ge.imati.cnr.it) and Andrea Giachetti (email: andrea.giachetti@univr.it) with the subject "SHREC17 track: Retrieval of surfaces with similar relief patterns". Then, an answer will be sent to each participant with the instructions to:

Preliminary results are now available. Please click here to download them.

Further Information

Important dates