SHREC'19 track: Feature Curve Extraction on Triangle Meshes

The aim of this SHREC'19 track is to evaluate the performance of automatic algorithms for the extraction of feature curves on triangle meshes. The goal is to identify where there are geometric lines that delineate a feature on the model. The track is organized by IMATI CNR . The SHREC'19 track: Feature Curve Extraction on Triangle Meshes final report will appear, after revision, in the proceedings of the EUROGRAPHICS Workshop on 3D Object Retrieval (3DOR).

Dataset, ground truth and evaluation

The dataset is created from collections od 3D models available online. The dataset contains triangulations, only. Each model has one connected component and can be with or without boundary. The challenge of the track is to highlight the vertices of the triangulation that delineate the feature/s on the model.

As an optional task, participants may look for similar feature curves across all those found on the dataset. In case of multiple features on the same model, the participants can group feature lines between those he found across all the models based on their shapes.

Registration and instructions for competitors
Each participant is requested to register to the track by sending an email to Elia Moscoso Thompson (email: elia.moscoso@ge.imati.cnr.it) with the subject SHREC'19 track: Feature Curve Extraction on Triangle Meshes. Then, an answer will be sent to each participant with the instructions to download the models so they can work on them. In more details:

• Models in the datasets are in ply format of different size and various vertex resolutions (up to 200000 vertices).
• All the models have at least one feature curve.
• Participants must provide by the 28th of February 2019:
  • The mandatory output(s). For each feature curve on each model, a plain ASCII file named modelnumber_featline.txt (e.g.: the first curve identified on the 2nd model must be named "2_1.txt") containing the indexes of the vertices that define the feature curve. Participants can resample/change file format or model representation (cloud points, voxels, segmentations... ) without restrictions, but the final results MUST be related to the original triangulation in ply format.

  • Optional Output. Participants may submit a similarity matrix of values ranging from 0 to 1, where the value 0 means that the models have identical feature lines while 1 excludes the similarity between feature lines (e.g.: 0 means that the feature lines that highlight eyes on two different models are identical).

  • The executables that generated the results. We accept Windows, Linux or Mac executables (scripts are also acceptable), that should be provided along with all the possible dependencies (.dll, .so, other executables...). Alternatively, the source code may be provided, together with full compilation instructions and their dependencies: we accept C, C++ and Matlab code. If a participant won't submit any executable, we will evaluate his/her run(s) but we will mention this fact in the final report.
• Participants can send 1 to 3 runs for the algorithm(s) they used.

Further Information

• The evaluations will be done automatically.
• The track results will be combined into a joint paper, to be published in the proceedings of the Eurographics Workshop on 3D Object Retrieval.
• The description of the track and its results will be presented at the Eurographics Workshop on 3D Object Retrieval (May 5-6, 2019).