Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
412039 | Neurocomputing | 2015 | 10 Pages |
3D shape retrieval is a fundamental task in many domains such as multimedia, graphics, CAD, and amusement. In this paper, we propose a 3D object retrieval approach by effectively utilizing low-level patches of 3D shapes, which are similar as superpixels in images. These patches are first obtained by means of stably over-segmenting 3D shape, and then we adopt five representative geometric features including shape diameter function, average geodesic distance, and heat kernel signature, to characterize these low-level patches. A large number of patches collected from shapes in a dataset are encoded into patch words by virtue of locality-constrained sparse coding under the consideration of local smooth sparsity. Input query is compared with 3D models in the dataset through probability distribution of patch words. Experiments reveal that the proposed method achieves comparable retrieval performance to state-of-the-art methods.