Repeatability of local reference frames for partial shape matching
- Glock (8 views - 7 MB)
- Fish (10 views - 8 MB)
- Amphora (14 views - 13 MB)
- Neptune (15 views - 22 MB)
- Buste (16 views - 3 MB)
- Dancing Children (26 views - 36 MB)
Surface matching deals with the ability of finding similarities between 3D surfaces and is a key task in scenarios such as 3D object recognition and surface registration. Last decade research on surface matching has been mainly focused on local rather than global approaches, for the former being able to withstand nuisances such as clutter and occlusions. Hence, research efforts have addressed the definition of local 3D descriptors, that is compact representations of surface points based on the characteristics of their neighborhood. Invariance to objects’ pose is an indispensable trait of every 3D descriptor. Most authors achieve it by using descriptions based on invariant Local Reference Frame (LRF) and on the description of the support with respect to such LRF coordinates. As long as the LRF turns out repeatable and robust to noise, the descriptor holds the potential for high distinctiveness since it can encode all the shape information within the support.
Fig. 1
Therefore, our first contribution is an extensive benchmark study whose purpose is to analyze and compare the repeatability of existing LRF proposals in a partial shape matching scenario, so as to elucidate on the methods most appropriate to this important task. Our experimental evaluation show that LRFs based on principal directions turn out unsuitable to partial shape matching mainly due to the local point density variations induced by the changes of the vantage point. On the other hand, the LRF associated with the Point Signatures descriptor reveals itself to be the most repeatable. As second contribution, we conceive and propose a novel LRF which requires the computation of surface normals only and deals with the problem of missing regions held by view borders that drastically decrease LRF repeatability. Experiments prove that the proposed LRF exhibits the highest repeatability and that its computation time is comparable to that of the fastest -and less repeatable- existing methods.
Dataset
It is possible to download the dataset with 9 models used for the experimental evaluation in [1]. Every ZIP archive contains a set of partial views in PLY format and a groundTruth.txt file with rototranslation matrices to align every partial view.
Three models belong to the Stanford 3D Scanning Repository:
Whereas six models belong to the AIM@SHAPE Repository:
Code
We can freely distribute an implementation of our LRF proposal for Visual C++ 2005 environment. To obtain the code, please write to (alioscia DOT petrelli AT unibo DOT it) and we will provide to send it to you.
Experimental Results
We make available the experimental results of the benchmark described in [1] among our and existing LRF proposals.
References
| [1] | A. Petrelli, L. Di Stefano, "On the repeatability of the local reference frame for partial shape matching", 13th International Conference on Computer Vision (ICCV), 2011. [PDF, supplementary material] |
