HSD: A 3D shape descriptor based on the Hilbert curve and a reduction dimensionality approach

Similarity searching based on 3D shape descriptors is an important process in content-based 3D shape retrieval tasks. The development of efficient 3D shape descriptors is still a challenge. This paper proposes a novel approach to characterize 3D shapes that is based on a Hilbert curve for scanning the volume, dimensionality reduction by discrete wavelet transform and artificial neural networks. Our proposal, called Hilbert based 3D-shape Description, yields a high level descriptor that preserves the relevant characteristics of a 3D shape. Our proposal is invariant under translation transformation and it is robust under scale transformation. The experiments were carried out using the Princeton Shape Benchmark. The evaluation of the results indicated a higher precision rate, when compared to the competitive works.