Non-rigid 3D shape classification using bag-of-feature techniques

Abstract

In this paper, we present a new method for 3D-shape categorization using Bag-of-Feature techniques (BoF). This method is based on vector quantization of invariant descriptors of 3D-object patches. We analyze the performance of two wellknown classifiers: the Naïve Bayes and the SVM. The results show the effectiveness of our approach and prove that the method is robust to non-rigid and deformable shapes, in which the class of transformations may be very wide due to the capability of such shapes to bend and assume different forms.

A Novel Local Surface Description for Automatic 3D Object Recognition in Low Resolution Cluttered Scenes

“…They then generated a 3D-tensor descriptor by constructing a local 3D grid over the range image, and summed the surface areas which intersected each bin of the 3D grid. In a method proposed by Tabia et al [19], a set of feature points was detected first. Then the geodesic distances were computed from each feature point to all the vertices on the 3D surface.…”

Iterative 3D shape classification by online metric learning

“…Most existing works (Csakany and Wallace, 2003;Biasotti et al, 2006;Huber et al, 2004;Tabia et al, 2011Tabia et al, , 2013Marini et al, 2011;Barra and Biasotti, 2014;Qin et al, 2014) have focused on learning a classifier by a large number of labeled samples to classify a given shape. As the training set utterly determines the scope of categories and the trained classifier is hardly updated, almost all of them cannot generalize well to unknown object category and multiple classification criterions to fit the needs of the diversity of taxonomies and the different application requirements.…”

Intrinsic spatial pyramid matching for deformable 3D shape retrieval

“…The codebook model has been shown to be a promising method for partial shape retrieval [8,9]. Recent efforts have also focused on finding the deformation invariance for nonrigid shapes by replacing the Euclidean metric with its geodesic counterpart [10]. The geodesic distance, however, suffers from strong sensitivity to topological noise, which limits its usefulness in real applications.…”